You are reading a research paper published on EntryLib as separate claims, each with its evidence. How to read this page

corrected · Version 3 · 18 November 2026 · Statistics and Probability

Corrected · 18 November 2026 at 13:08 UTC

Version 3 answers a reader's comments. Corrected: C3 no longer says sMAPE is minimised by the median for small errors, only near it, with a bound; C4 keeps the weighted median, and C12 now holds what was said about the ordinary median; the measure called msMAPE in version 2 is the Suilin loss. Added: how well a test set tells two forecasts apart as the actuals spread out (C18); the large-sample distribution of the sMAPE minimiser (C19); Smyl's msMAPE (C20) and what it rewards on intermittent demand (C21). Also added: sections, the definition of sMAPE first, and a limitation on C6.

By Ivan Svetunkov (ORCID 0000-0001-7826-0281)

What does sMAPE reward? An analytical and simulation study

Responsible researcher: Ivan Svetunkov (0000-0001-7826-0281)

CC-BY-4.0 · Keywords: forecast evaluation, error measures, sMAPE, point forecasts, consistent scoring functions, proof, simulation study, msMAPE, Suilin loss, intermittent demand

Contributions
  • Claude Code · derivations and simulations (AI agent): Methodology, Formal analysis, Software, Investigation, Visualization, Writing – original draft
  • Ivan Svetunkov (0000-0001-7826-0281): Conceptualization, Supervision, Validation, Writing – review & editing
  • Claude Code · version 3: corrections and new derivations (AI agent): Methodology, Formal analysis, Software, Investigation, Visualization, Writing – original draft

Made with claimset-lab.

Text file for AI

Text file: the whole paper as plain text, to attach when your assistant cannot open the address.

The message

Read https://entrylib.org/p/[email protected], a research paper written as claims with their evidence. Give me a brief summary of its main findings and how well each is supported. If you cannot open the link, tell me instead of guessing. Then wait for my questions.

Summary

Which forecast does sMAPE reward? sMAPE has been criticised for its asymmetry and its behaviour near zero, but the forecast that minimises it had been found only in special cases. We show that for positive values sMAPE depends only on the ratio of actual to forecast, not on their difference: it is the hyperbolic tangent of half the absolute log error. So it scores a forecast k times too high and one k times too low the same, and is asymmetric only for equal absolute errors. Being bounded, it has a finite expected value for every distribution. Its target is a median of the actuals, each weighted by how close it is to the forecast by ratio. This is the median when the log of the actual is symmetric and unimodal; otherwise it can move to the side of the median that carries more weight. With two well-separated groups of actuals, the median becomes a local maximum. On intermittent demand the target is 0 once the share of zeros passes a threshold below one half. Otherwise sMAPE aims at a weighted median of the non-zero actuals. As the actuals spread out, expected sMAPE approaches its bound for every forecast, so forecasts become hard to tell apart, and the target moves from the median to the mode of the log. Where sMAPE and MAE both target the median, a test set's sMAPE minimiser estimates it less precisely than the sample median, increasingly so as the actuals spread out.

Highlights

  • SupportS 100For y>0y > 0 and f>0f > 0, the sMAPE term s(y,f)=|y−f|/(|y|+|f|)s(y,f) = |y-f|/(|y|+|f|) depends on yy and ff only through their ratio r=y/fr = y/f: s=|r−1|/(r+1)s = |r - 1|/(r + 1). C1
  • SupportS 100Let the actual yy be positive and take no single value with positive probability, let mm be its ordinary (unweighted) median, and let its expected sMAPE have a single local minimum. With the forecast at mm, compare the total weight w(y,m)=y/(y+m)2w(y,m) = y/(y+m)^2 of the actuals above mm with that of the actuals below it. (a) If the actuals above mm carry more weight, the sMAPE target lies above mm; if those below carry more, it lies below mm. (b) If the two weights are equal, mm is a point where expected sMAPE is flat. It is then the target if, in addition, 2g(ln⁡m)>E[sech2⁡(ln⁡y−ln⁡m2)tanh⁡(|ln⁡y−ln⁡m|2)],2g(\ln m) > \mathrm E\Big[\operatorname{sech}^2\Big(\frac{\ln y - \ln m}{2}\Big)\tanh\Big(\frac{|\ln y - \ln m|}{2}\Big)\Big], where gg is the density of ln⁡y\ln y, continuous at ln⁡m\ln m: this makes mm a strict local minimum. Without it, mm can be a flat point of inflection, with the target elsewhere. (c) When ln⁡y\ln y has a density symmetric about ln⁡m\ln m and unimodal, mm is the only target; lognormal and log-Laplace actuals are examples. C12
  • SupportS 100Let the actual yy be positive and take no single value with positive probability. A forecast ff has the lowest expected sMAPE only if the actuals below ff and those above it carry equal total weight w(y,f)=y/(y+f)2w(y,f) = y/(y+f)^2: E[w(y,f)𝟏{y<f}]=E[w(y,f)𝟏{y>f}].\mathrm E[w(y,f)\,\mathbf 1\{y<f\}] = \mathrm E[w(y,f)\,\mathbf 1\{y>f\}]. So the sMAPE target is a median of the actuals weighted by ww, where the ordinary median gives every actual the same weight. An actual's weight is largest when it equals the forecast, is the same for y/f=ry/f = r and 1/r1/r, and falls as y/fy/f moves away from 1 in either direction. C4
  • SupportS 89Where sMAPE and MAE both target the median (log-symmetric outcomes), the sMAPE minimiser of a test set varies more than the sample median for every spread and test-set size studied: its relative efficiency is below 1 and falls as the actuals spread out, at n=10000n = 10000 from 0.97 to 0.03 on lognormal actuals as σ\sigma grows from 0.25 to 6, and from 1.00 to 0.34 on log-Laplace actuals as bb grows from 0.25 to 6, close to the large-sample values. C15
  • SupportS 94Let ln⁡y\ln y have two modes, at ln⁡m−a\ln m - a and ln⁡m+a\ln m + a (a>0a > 0), with the actuals concentrated near them: a mixture of a distribution near ln⁡m−a\ln m - a and its mirror image about ln⁡m\ln m, with weights pp and 1−p1 - p. (a) With equal weights the median mm lies in the gap between the two groups; when the groups are narrow enough it has a higher expected sMAPE than forecasts near either mode, and it is a local maximum of expected sMAPE; there is a minimiser on each side of the median, and these approach me−am\,e^{-a} and meam\,e^{a} as the groups narrow. (b) With unequal weights and narrow groups, the best forecast is near the mode of the larger group. C6

Context

Extends Tofallis (2015) · from C1
Tofallis (2015) lists the sMAPE term, |y-f|/[(|y|+|f|)/2], among Tornqvist et al.'s measures of relative change; C1 shows it depends on the actual and the forecast only through their ratio.
Qualifies Svetunkov (2026a), C33 · from C2
C33 says sMAPE penalises a forecast below the outcome more than one above it. That holds for two forecasts the same absolute distance away (C2 b); a forecast k times too low and one k times too high score the same (C2 a).
Refines Goodwin and Lawton (1999) · from C2
Goodwin and Lawton (1999) showed that sMAPE is asymmetric: for the same absolute error it scores a forecast above the actual better than one below. C2 proves this (b) and shows that by ratio sMAPE is symmetric (a).
Refines Hyndman and Koehler (2006) · from C2
Hyndman and Koehler (2006) note that for the same actual sMAPE penalises low forecasts more heavily than high ones; C2 shows this holds for equal absolute errors and not for equal ratios.
Refines Hewamalage et al. (2023) · from C2
Hewamalage et al. (2023) note that sMAPE penalises underestimates more than overestimates for the same actual; C2 shows this holds for equal absolute errors and not for equal ratios.
Refines Chen et al. (2017) · from C2
Chen et al. (2017) show a heavier penalty on a forecast 10% under the actual than on one 10% over; C2 shows that forecasts the same factor too high and too low score the same.
Extends Tofallis (2015) · from C3
Tofallis (2015) relates sMAPE to the log accuracy ratio among relative measures; C3 shows the sMAPE term is exactly the hyperbolic tangent of half the absolute log ratio.
Explains Svetunkov (2026b), C5 · from C4
C5 found sMAPE's preferred candidate close to the median on other distributions; C4 shows why: the target is a weighted median, on the side of the median that carries more weight (C12).
Addresses Tofallis (2015) · from C4
Tofallis (2015) states that minimising sMAPE does not produce an established measure of location; C4 identifies the target as a median weighted by y/(y+f)^2.
Qualifies Hewamalage et al. (2023) · from C4
Hewamalage et al. (2023) say sMAPE may tend towards selecting a slightly overestimating model; C4 shows the target is a weighted median, which lies on the side of the median that carries more weight, above or below it (C12).
Replicates Anderson (1955) · from C5
C5 is an elementary case of Anderson's theorem on symmetric unimodal densities, proved here directly for the expected sMAPE.
Refines Svetunkov (2026b), C6 · from C7
C6 found sMAPE preferring zero on counts with more than half zeros; C7 gives the exact threshold, which is below one half whenever the positive outcomes vary.
Extends Kolassa (2026) · from C7
Kolassa (2026) shows that sMAPE elicits a zero forecast for sufficiently intermittent actuals; C7 gives the share of zeros at which it does, which is below one half.
Extends Hewamalage et al. (2023) · from C9
Hewamalage et al. (2023) note that sMAPE is bounded by 200; C9 adds that its expected value is therefore finite for every distribution and that a best forecast exists.
Supports Svetunkov (2026a), C6 · from C10
C6 recommends against sMAPE for point forecasts; C10 shows that on widely spread actuals every forecast scores close to the worst value and the target moves to the mode of ln y.
Supports Kolassa (2026) · from C10
Kolassa (2026) concludes that sMAPE should not be used; C10 shows that on widely spread actuals every forecast scores close to the worst value.
Explains Svetunkov (2026b), C5 · from C12
C5 found by simulation that sMAPE prefers the median on lognormal and log-Laplace outcomes; C12 proves it for every symmetric unimodal ln y.
Supports Svetunkov (2026b), C11 · from C15
C11 found sMAPE's choices in short test sets spreading more than MAE's; C15 measures the loss of precision for test sets of 6 to 10000 outcomes.
Supports Svetunkov (2026a), C6 · from C15
C6 recommends against sMAPE for point forecasts; C15 shows that even where it targets the median it estimates it less precisely than the sample median.
Supports Kolassa (2026) · from C15
Kolassa (2026) concludes that sMAPE should not be used; C15 shows that even where it targets the median it estimates it less precisely than the sample median.
Extends Hewamalage et al. (2021) · from C17
Hewamalage, Bergmeir and Bandara (2021) evaluate forecasts with the Suilin loss, Suilin's modification of sMAPE for series with values near zero; C17 shows that it keeps sMAPE's behaviour away from zero and becomes an absolute error near zero.
Supports Svetunkov (2026a), C6 · from C18
C6 recommends against sMAPE for point forecasts; C18 shows that on widely spread actuals a test set tells two forecasts apart less and less well under sMAPE.
Supports Kolassa (2026) · from C18
Kolassa (2026) concludes that sMAPE should not be used; C18 shows that on widely spread actuals a test set tells two forecasts apart less and less well under sMAPE.
Extends Kolassa (2026) · from C20
Kolassa (2026, Proposition 3) shows that the minimisers of expected msMAPE on counts are integers; C20 shows where the msMAPE term equals the sMAPE term and that it removes the jump at the zero forecast.
Extends Smyl (2025) · from C20
Smyl (2025) proposed msMAPE, the sparse-proof sMAPE, as Kolassa (2026) reports; Smyl's article was not read. C20 states properties of its term as Kolassa (2026, Definition 4) defines it.

Definitions

Methods 4

D16 selection design: For each outcome distribution: derive the candidate forecasts and their expected scores; then, for each test-set size nn, draw test sets of nn independent outcomes, score every candidate by every measure on each test set, and count the candidate with the lowest score as the measure's choice (a tie of jj candidates counts 1/j1/j each; a test set in which every candidate scores infinite chooses none).

D17 efficiency design: For each outcome distribution and test-set size nn: draw test sets of nn independent outcomes; in each, compute the sample median and the exact sMAPE minimiser (an outcome of the test set, since the test set's sMAPE is concave in ln⁡f\ln f between consecutive outcomes); over test sets, compare the variances of their logarithms.

D18 flatness design: For families whose spread σ\sigma of ln⁡y\ln y grows from 0.1 to 10: derive the sMAPE target, the expected sMAPE of forecasts at fixed quantile levels and the large-spread approximation of C10; on test sets, count how often each measure scores the target lower than a forecast away from it, and where the test set's sMAPE minimiser falls.

D31 formula checks: Numerical checks of the formulas of the formal claims (claimset-code/smape-target/check_formulas.py, seed 6; check_jump.py for C11): identities and inequalities on 200000 random pairs, derivatives against finite differences, minimisers, balances and thresholds against quadrature and bisection on every setting of the selection design, and the large-spread expansion against its bound. For the claims new or reworded in version 3 (C3, C4, C12, C17 to C20): check_v3.py (seed 8), check_h39.py and check_corrected.py (seed 8), in the same way.

Variables and measures 15

D2 sMAPE: Symmetric mean absolute percentage error of a point forecast ff over a test set of actuals y1,…,yny_1, \dots, y_n: 200n∑t|yt−f|/(|yt|+|f|)\frac{200}{n}\sum_t |y_t - f|/(|y_t|+|f|), where a term with yt=f=0y_t = f = 0 counts as 0. Introduced as the symmetric MAPE by Makridakis (1993), whose denominator had no absolute values; for positive actuals and forecasts the two agree. Source: Makridakis (1993).

D1 sMAPE term: The contribution of one actual yy and one point forecast ff to sMAPE: s(y,f)=|y−f|/(|y|+|f|)s(y,f) = |y-f|/(|y|+|f|), between 0 and 1, with s=0s = 0 when y=fy = f (including y=f=0y = f = 0, where the fraction is 0/00/0).

D3 expected sMAPE: The expected sMAPE of a fixed forecast ff under the distribution of the actual yy: 200E[s(y,f)]200\,\mathrm{E}[s(y, f)], where E\mathrm E is the expectation over the random actual yy. Computed from the distribution (by integration or exact sums), never from a sample.

D4 log error: For a positive actual yy and a positive forecast ff, z=ln⁡y−ln⁡fz = \ln y - \ln f.

D5 weight: For a positive actual yy and a positive forecast ff, w(y,f)=y/(y+f)2w(y,f) = y/(y+f)^2: how much the actual pulls the forecast that minimises expected sMAPE, since the derivative of E[s(y,f)]\mathrm E[s(y,f)] in ff is 2E[w(y,f)(𝟏{y<f}−𝟏{y>f})]2\,\mathrm E[w(y,f)(\mathbf 1\{y<f\} - \mathbf 1\{y>f\})]. Largest at y=fy = f, the same for y/f=ry/f = r and 1/r1/r.

D10 RMSE: Root mean squared error of a point forecast ff over a test set of actuals y1,…,yny_1, \dots, y_n: 1n∑t(yt−f)2\sqrt{\frac1n\sum_t (y_t-f)^2}.

D11 MAE: Mean absolute error of a point forecast ff over a test set of actuals y1,…,yny_1, \dots, y_n: 1n∑t|yt−f|\frac1n\sum_t |y_t-f|.

D12 MAPE: Mean absolute percentage error of a point forecast ff over a test set of actuals y1,…,yny_1, \dots, y_n: 100n∑t|yt−f|/|yt|\frac{100}{n}\sum_t |y_t-f|/|y_t|; infinite when an actual is zero.

D19 log-MAE: 100 times the mean absolute error of ln⁡f\ln f as a forecast of ln⁡y\ln y: 100n∑t|ln⁡yt−ln⁡f|\frac{100}{n}\sum_t |\ln y_t - \ln f|. Unbounded; sMAPE's small-error limit (C3); minimised by the median.

D28 Suilin term: The contribution of one actual yy and one forecast ff to the Suilin loss: s~(y,f)=|y−f|max⁡(|y|+|f|+δ,0.5+δ),δ=0.1;\tilde s(y,f) = \frac{|y-f|}{\max(|y|+|f|+\delta,\ 0.5+\delta)}, \qquad \delta = 0.1; the Suilin loss is 200 times its mean over a test set. It replaces the denominator |y|+|f||y|+|f| of the sMAPE term. Proposed by Suilin (2017) in code for a Kaggle competition, as a smoothed variant of sMAPE; defined in print by Hewamalage, Bergmeir and Bandara (2021, Eq. 15), who write the constant δ\delta as ϵ\epsilon and the factor 2 into the measure differently (a constant factor changes no property here). Hewamalage, Ackermann and Bergmeir (2023) call it msMAPE; Smyl (2025) gives that name to a different measure, so this study follows Kolassa (2026) and calls it the Suilin loss. Version 2 of this paper called it msMAPE. Source: Hewamalage et al. (2021).

D33 msMAPE term: The contribution of one actual y≥0y \ge 0 and one forecast f≥0f \ge 0 to the sparse-proof sMAPE (msMAPE) of Smyl (2025), as Kolassa (2026, Definition 4) states it, for a small constant ν>0\nu > 0 (Kolassa writes it mm, which is the median here): sν(y,f)=|f−y|f+max⁡{y,ν},s_\nu(y,f) = \frac{|f-y|}{f+\max\{y, \nu\}}, which is 0 when y=f=0y = f = 0; msMAPE is 200 times its mean over a test set (Kolassa writes the factor 2 into the term). It replaces the actual in the denominator of the sMAPE term by ν\nu when the actual is below ν\nu. Source: Kolassa (2026).

D8 sMAPE minimiser of a test set: The forecast that minimises the sMAPE of one test set y1,…,yny_1, \dots, y_n, f^n=arg⁡minf⁡1n∑ts(yt,f)\hat f_n = \arg\min_f \frac1n\sum_t s(y_t, f), found numerically (the sample counterpart of the sMAPE target).

D9 relative efficiency: For a test-set size nn and an outcome distribution where sMAPE and MAE both target the median: RE(n)=Var⁡(ln⁡m^n)/Var⁡(ln⁡f^n)\mathrm{RE}(n) = \operatorname{Var}(\ln \hat m_n) / \operatorname{Var}(\ln \hat f_n) over test sets, where m^n\hat m_n is the sample median (the MAE minimiser) and f^n\hat f_n the sMAPE minimiser of the same test set. Above 1, sMAPE's minimiser varies less; below 1, MAE's does.

D20 ranking agreement: For one distribution, one measure, a test-set size nn, the measure's target and a forecast at another quantile level: the probability that a test set of nn outcomes gives the target the strictly lower score (ties count as disagreement); 0.5 is a coin toss.

D32 signal-to-noise ratio of a comparison: For a measure with term L(y,f)L(y,f), two forecasts f1f_1 and f2f_2 and a distribution of the actual yy: |E[D]|/sd(D)|\mathrm E[D]|/\mathrm{sd}(D), where D=L(y,f2)−L(y,f1)D = L(y,f_2) - L(y,f_1) is the difference of the two forecasts' scores on one actual. Over a test set of nn independent actuals the mean of DD is n\sqrt n times this many of its standard errors from zero: the larger the ratio, the smaller the test set that tells the two forecasts apart.

Populations 9

D21 inverse Gaussian outcomes: Inverse Gaussian outcomes with mean 100 and shape 1000, 200, 100, 50, 20 or 10 (coefficient of variation 0.32 to 3.16): positive, skewed right, every moment and E[1/y]\mathrm E[1/y] finite.

D22 log-skewed outcomes: y=100exy = 100\,e^x with x=±σ(G−α)/αx = \pm\sigma(G-\alpha)/\sqrt\alpha, G∼Gamma(α,1)G \sim \mathrm{Gamma}(\alpha, 1): ln⁡y\ln y with standard deviation σ\sigma (0.5 or 1; sd in the settings' names) and skewness ±2/α\pm 2/\sqrt\alpha (-2, -1, -0.5, +0.5, +1, +2), and the lognormal of the same σ\sigma (skewness 0). Some have infinite moments: E[y]\mathrm E[y] infinite (σ=1\sigma = 1, skewness +2), E[y2]\mathrm E[y^2] infinite (σ=0.5\sigma = 0.5, skewness +2; σ=1\sigma = 1, skewness +1), E[1/y]\mathrm E[1/y] infinite (σ=1\sigma = 1, skewness -2).

D23 bimodal outcomes: y=100exy = 100\,e^x, xx normal with mean −1-1 (probability pp) or +1+1, standard deviation 0.25; p=0.5p = 0.5 (equal modes), 0.65 (left mode higher), 0.35 (right mode higher).

D24 outcomes straddling zero: Normal outcomes with standard deviation 1 and mean 0, 0.5 or 1: 50%, 31% and 16% negative; E[1/|y|]\mathrm E[1/|y|] infinite.

D25 zero-inflated outcomes: Lognormal outcomes (median 100, σ\sigma 0.5 or 1) replaced by 0 with probability p0p_0: 0.25, 0.42, 0.47, 0.6 for σ=0.5\sigma = 0.5 and 0.25, 0.37, 0.42, 0.6 for σ=1\sigma = 1, on both sides of the zero threshold of C7 (0.4469 and 0.3940).

D26 log-symmetric outcomes: Lognormal outcomes with σ\sigma, and log-Laplace outcomes with scale bb, each 0.25, 0.5, 1, 1.5, 2, 3, 4 or 6, median 100: ln⁡y\ln y symmetric and strictly unimodal, so that sMAPE and MAE both target the median (C5).

D27 widening outcomes: ln⁡y=ln⁡100+σε\ln y = \ln 100 + \sigma\varepsilon with ε\varepsilon standard normal (lognormal), standard Gumbel (log-Gumbel right, ln⁡y\ln y skewed right) or minus standard Gumbel (log-Gumbel left), for σ\sigma = 0.1, 0.25, 0.5, 1, 1.5, 2, 3, 4, 6 and 10.

D30 Poisson counts: Poisson outcomes with means 0.3, 0.5, 1, 1.5, 3 and 10: counts whose share of zeros falls from 0.74 to almost 0.

D35 intermittent demand settings: Demand with many zeros. Counts: Poisson with means 0.1, 0.3, 0.5, 0.7, 1, 1.5 and 3 (share of zeros from 0.90 to 0.05) and, overdispersed, negative binomial with size 0.5 and means 0.5, 1 and 2 (share of zeros 0.71, 0.58 and 0.45). Non-integer demand: lognormal demand sizes (median 1, σ=1\sigma = 1) replaced by 0 with probability 0.2, 0.4, 0.6 or 0.8.

Concepts 7

D6 sMAPE target: The value or values of ff that minimise expected sMAPE for a given distribution of the actual yy, arg⁡minf⁡E[s(y,f)]\arg\min_f \mathrm E[s(y,f)]: the functional sMAPE rewards. It may be a single value or several.

D7 quantile profile: For an outcome distribution with quantile function QQ and a forecast ff, the pull of the outcome at each quantile level uu, u↦w(Q(u),f)u \mapsto w(Q(u), f) on (0,1)(0,1); the minimiser's quantile level balances its integrals below and above.

D29 Suilin target: The value or values of ff that minimise the expected Suilin term for a given distribution of the actual: the functional the Suilin loss rewards, as the sMAPE target is the one sMAPE rewards.

D34 msMAPE target: The value or values of ff that minimise the expected msMAPE term for a given distribution of the actual and a given constant ν\nu: the functional msMAPE rewards, as the sMAPE target is the one sMAPE rewards.

D13 candidate forecast: A point forecast fixed from the outcome distribution before any test set is drawn and the same for every test set: a quantile at a level from 0.01 to 0.99, the mean, the forecast 0, the log-mode, the minimiser of expected MAPE, or a local minimiser of expected sMAPE. It is never estimated from a sample.

D14 preferred candidate: Of the candidate forecasts for one distribution, the one a measure gives the lowest expected score, or, where every expected score is infinite but their differences are finite, the one the differences rank first (C8); none where no forecast is ranked first.

D15 derived value: A value computed from the outcome distribution itself by numerical integration of its density (Gauss-Legendre quadrature and a convolution on a grid of 524288 points of ln⁡y\ln y), never from a sample: the expected scores, the local minima of expected sMAPE, the formulas of the hypotheses.

Principal claims

SupportS 100C1 · formal · planned test · novelty: extends earlier work, by the authors

For y>0y > 0 and f>0f > 0, the sMAPE term s(y,f)=|y−f|/(|y|+|f|)s(y,f) = |y-f|/(|y|+|f|) depends on yy and ff only through their ratio r=y/fr = y/f: s=|r−1|/(r+1)s = |r - 1|/(r + 1).

SupportS 100C3 · formal · planned test · novelty: extends earlier work, by the authors

For y>0y > 0 and f>0f > 0, the sMAPE term is s=tanh⁡(|ln⁡y−ln⁡f|/2)s = \tanh(|\ln y - \ln f|/2) (log error ln⁡y−ln⁡f\ln y - \ln f). (a) It is 0 when y=fy = f, strictly increasing and concave in |ln⁡y−ln⁡f||\ln y - \ln f|, and bounded by 1, which it approaches as |ln⁡y−ln⁡f||\ln y - \ln f| grows. (b) 0≤12|ln⁡y−ln⁡f|−s≤124|ln⁡y−ln⁡f|30 \le \tfrac12|\ln y - \ln f| - s \le \tfrac1{24}|\ln y - \ln f|^3; so when all log errors are small, sMAPE is about 100 times the mean absolute error of ln⁡f\ln f as a forecast of ln⁡y\ln y (log-MAE). (c) The log-MAE of a test set is minimised by its median; sMAPE is then minimised near the median, not in general at it. For positive actuals y1,…,yny_1, \dots, y_n with a median mm, a forecast f*f^* that minimises their sMAPE satisfies 1n∑t|ln⁡yt−ln⁡f*|−1n∑t|ln⁡yt−ln⁡m|≤112⋅1n∑t|ln⁡yt−ln⁡f*|3:\frac1n\sum_t|\ln y_t - \ln f^*| - \frac1n\sum_t|\ln y_t - \ln m| \le \frac1{12}\cdot\frac1n\sum_t|\ln y_t - \ln f^*|^3: its mean absolute log error exceeds the smallest possible by at most a twelfth of its mean cubed absolute log error. The minimiser is the median itself only under further conditions on the distribution of the actuals.

SupportS 100C4 · formal · planned test · novelty: new, by the authors

Let the actual yy be positive and take no single value with positive probability. A forecast ff has the lowest expected sMAPE only if the actuals below ff and those above it carry equal total weight w(y,f)=y/(y+f)2w(y,f) = y/(y+f)^2: E[w(y,f)𝟏{y<f}]=E[w(y,f)𝟏{y>f}].\mathrm E[w(y,f)\,\mathbf 1\{y<f\}] = \mathrm E[w(y,f)\,\mathbf 1\{y>f\}]. So the sMAPE target is a median of the actuals weighted by ww, where the ordinary median gives every actual the same weight. An actual's weight is largest when it equals the forecast, is the same for y/f=ry/f = r and 1/r1/r, and falls as y/fy/f moves away from 1 in either direction.

SupportS 94C6 · formal · planned test · novelty: new, by the authors

Let ln⁡y\ln y have two modes, at ln⁡m−a\ln m - a and ln⁡m+a\ln m + a (a>0a > 0), with the actuals concentrated near them: a mixture of a distribution near ln⁡m−a\ln m - a and its mirror image about ln⁡m\ln m, with weights pp and 1−p1 - p. (a) With equal weights the median mm lies in the gap between the two groups; when the groups are narrow enough it has a higher expected sMAPE than forecasts near either mode, and it is a local maximum of expected sMAPE; there is a minimiser on each side of the median, and these approach me−am\,e^{-a} and meam\,e^{a} as the groups narrow. (b) With unequal weights and narrow groups, the best forecast is near the mode of the larger group.

SupportS 100C7 · formal · planned test · novelty: extends earlier work, by the authors

A forecast of the opposite sign to the actual, a zero forecast of a non-zero actual, and a non-zero forecast of a zero actual all give the sMAPE term its largest value, 1. Let the actual yy be 0 with probability p0p_0 and otherwise positive, and s+s_+ the lowest expected ss a positive forecast reaches on the positive actuals. The forecast 0 is the sMAPE target when p0>(1−s+)/(2−s+)p_0 > (1 - s_+)/(2 - s_+), and a positive forecast does when the inequality is reversed; the threshold is below 1/21/2 whenever s+>0s_+ > 0.

SupportS 100C9 · formal · planned test · novelty: extends earlier work, by the authors

For actuals and forecasts of any sign, 0≤s≤10 \le s \le 1. So the expected sMAPE of every forecast is finite, between 0 and 200, whatever the distribution of the actuals and without any moment of it being finite (unlike RMSE, MAE and MAPE, C8). For a positive random actual yy, at least one forecast has the lowest expected sMAPE.

SupportS 100C10 · formal · planned test · novelty: new, by the authors

Let ln⁡y=μ+σε\ln y = \mu + \sigma\varepsilon, where ε\varepsilon has a fixed density hh with bounded continuous first and second derivatives, and write a forecast as f=eμ+σcf = e^{\mu + \sigma c} (its quantile level, the probability that ε≤c\varepsilon \le c, does not depend on σ\sigma). As σ\sigma grows, E[s(y,f)]=1−4ln⁡2h(c)/σ+O(σ−3)\mathrm E[s(y,f)] = 1 - 4\ln 2\,h(c)/\sigma + O(\sigma^{-3}) uniformly in cc. (a) The expected sMAPE of every forecast tends to its bound 200, and the difference between forecasts at two fixed quantile levels shrinks like 1/σ1/\sigma. (b) When hh has a single highest point c0c_0, the quantile level of the sMAPE target tends to that of c0c_0, the mode of ln⁡y\ln y, which differs from 1/2 when the mode of ε\varepsilon is not its median. (c) As σ→0\sigma \to 0, when ε\varepsilon has a unique median and a finite E|ε|3\mathrm E|\varepsilon|^3, the quantile level of the target tends to 1/2: sMAPE targets the median for narrow distributions and the mode of ln⁡y\ln y for wide ones.

SupportS 100C12 · formal · planned test · novelty: new, by the authors

Let the actual yy be positive and take no single value with positive probability, let mm be its ordinary (unweighted) median, and let its expected sMAPE have a single local minimum. With the forecast at mm, compare the total weight w(y,m)=y/(y+m)2w(y,m) = y/(y+m)^2 of the actuals above mm with that of the actuals below it. (a) If the actuals above mm carry more weight, the sMAPE target lies above mm; if those below carry more, it lies below mm. (b) If the two weights are equal, mm is a point where expected sMAPE is flat. It is then the target if, in addition, 2g(ln⁡m)>E[sech2⁡(ln⁡y−ln⁡m2)tanh⁡(|ln⁡y−ln⁡m|2)],2g(\ln m) > \mathrm E\Big[\operatorname{sech}^2\Big(\frac{\ln y - \ln m}{2}\Big)\tanh\Big(\frac{|\ln y - \ln m|}{2}\Big)\Big], where gg is the density of ln⁡y\ln y, continuous at ln⁡m\ln m: this makes mm a strict local minimum. Without it, mm can be a flat point of inflection, with the target elsewhere. (c) When ln⁡y\ln y has a density symmetric about ln⁡m\ln m and unimodal, mm is the only target; lognormal and log-Laplace actuals are examples.

SupportS 89C15 · descriptive · planned estimate · novelty: new, by the authors

Where sMAPE and MAE both target the median (log-symmetric outcomes), the sMAPE minimiser of a test set varies more than the sample median for every spread and test-set size studied: its relative efficiency is below 1 and falls as the actuals spread out, at n=10000n = 10000 from 0.97 to 0.03 on lognormal actuals as σ\sigma grows from 0.25 to 6, and from 1.00 to 0.34 on log-Laplace actuals as bb grows from 0.25 to 6, close to the large-sample values.

SupportS 100C17 · formal · planned test · novelty: extends earlier work, by the authors

The Suilin term has the properties of the sMAPE term except near zero. (a) Where |y|+|f|≥0.5|y|+|f| \ge 0.5, s~(y,f)=s(y,f)|y|+|f||y|+|f|+δ:\tilde s(y,f) = s(y,f)\,\frac{|y|+|f|}{|y|+|f|+\delta}: the sMAPE term shrunk by a factor between 5/65/6 and 1 that tends to 1 as |y|+|f||y|+|f| grows. So for actuals and forecasts large compared with δ\delta, s~\tilde s is close to ss, though no longer exactly a function of the ratio y/fy/f; on counts, where every actual is 0 or at least 1, this case covers every pair except a zero actual with |f|<0.5|f| < 0.5. (b) Where |y|+|f|<0.5|y|+|f| < 0.5, s~(y,f)=|y−f|/(0.5+δ)\tilde s(y,f) = |y-f|/(0.5+\delta), an absolute error: a zero forecast of a zero actual scores 0, and a forecast ff with |f|<0.5|f| < 0.5 of a zero actual scores |f|/(0.5+δ)|f|/(0.5+\delta), where the sMAPE term gives every non-zero forecast of a zero actual its largest value, 1. (c) 0≤s~<10 \le \tilde s < 1 everywhere, so the expected Suilin loss of every forecast is finite.

SupportS 100C18 · formal · planned test · novelty: new, by the authors

Let ln⁡y=μ+σε\ln y = \mu + \sigma\varepsilon, where ε\varepsilon has a fixed density hh with bounded continuous first and second derivatives and distribution function HH, and compare two forecasts at fixed quantile levels, f1=eμ+σc1f_1 = e^{\mu+\sigma c_1} and f2=eμ+σc2f_2 = e^{\mu+\sigma c_2} with c1<c2c_1 < c_2, by their signal-to-noise ratio of a comparison. (a) For the sMAPE term, as σ\sigma grows, E[D]=4ln⁡2(h(c1)−h(c2))σ+O(σ−3),Var⁡(D)=4(2ln⁡2−1)(h(c1)+h(c2))σ+O(σ−2).\mathrm E[D] = \frac{4\ln 2\,(h(c_1) - h(c_2))}{\sigma} + O(\sigma^{-3}), \qquad \operatorname{Var}(D) = \frac{4(2\ln 2 - 1)(h(c_1)+h(c_2))}{\sigma} + O(\sigma^{-2}). So when h(c1)≠h(c2)h(c_1) \ne h(c_2) the ratio is 2ln⁡22ln⁡2−1|h(c1)−h(c2)|h(c1)+h(c2)σ−1/2(1+O(1/σ)):\frac{2\ln 2}{\sqrt{2\ln 2 - 1}}\,\frac{|h(c_1)-h(c_2)|}{\sqrt{h(c_1)+h(c_2)}}\,\sigma^{-1/2}\,(1 + O(1/\sigma)): it falls to zero like σ−1/2\sigma^{-1/2}, and the size of the test set that keeps the mean difference a given number of standard errors from zero grows in proportion to σ\sigma. (b) For the absolute log error |ln⁡y−ln⁡f||\ln y - \ln f| (log-MAE), D=σ(|ε−c2|−|ε−c1|)D = \sigma(|\varepsilon - c_2| - |\varepsilon - c_1|), so the ratio, wherever sd(D)>0\mathrm{sd}(D) > 0, does not depend on σ\sigma. (c) For the absolute error |y−f||y - f| (MAE), as σ\sigma grows the ratio tends to |2H(c2)−1|2H(c2)(1−H(c2))\frac{|2H(c_2) - 1|}{2\sqrt{H(c_2)(1 - H(c_2))}} when 0<H(c2)<10 < H(c_2) < 1; the limit is positive unless the higher forecast is the median. (d) For the squared error (y−f)2(y-f)^2 (RMSE), when E[y2]\mathrm E[y^2] is finite the ratio is |2E[y]−f1−f2|/(2sd(y))|2\,\mathrm E[y] - f_1 - f_2|/(2\,\mathrm{sd}(y)); for lognormal actuals, eσ2/2e^{\sigma^2/2} times the ratio tends to 1 as σ\sigma grows. For widely spread actuals, then: the absolute log error keeps its power to tell two forecasts apart wherever its ratio is defined. The absolute error keeps it when 0<H(c2)<10 < H(c_2) < 1 and the higher forecast is not the median. sMAPE loses it at the rate σ−1/2\sigma^{-1/2} when h(c1)≠h(c2)h(c_1) \ne h(c_2). It loses it faster, with a ratio of at most order σ−5/2\sigma^{-5/2}, when h(c1)=h(c2)>0h(c_1) = h(c_2) > 0. The squared error on lognormal actuals loses it faster than any power of σ\sigma.

SupportS 100C19 · formal · planned test · novelty: new, by the authors

Let ln⁡y\ln y have a density gg that is symmetric about ln⁡m\ln m, unimodal (g(ln⁡m+v1)≥g(ln⁡m+v2)g(\ln m + v_1) \ge g(\ln m + v_2) whenever |v1|≤|v2||v_1| \le |v_2|, strictly on a set of positive probability) and continuous at ln⁡m\ln m. Write x=ln⁡y−ln⁡mx = \ln y - \ln m, A=E[sech2⁡(x/2)tanh⁡(|x|/2)]A = \mathrm E[\operatorname{sech}^2(x/2)\tanh(|x|/2)] and B=E[sech4⁡(x/2)]B = \mathrm E[\operatorname{sech}^4(x/2)]. (a) A<2g(ln⁡m)A < 2g(\ln m). (b) Over test sets of nn independent actuals, the sMAPE minimiser of a test set f^n\hat f_n (a forecast with the lowest sMAPE of the test set) tends to mm in probability, and n(ln⁡f^n−ln⁡m)\sqrt n(\ln\hat f_n - \ln m) is asymptotically normal with mean 0 and variance B/(2g(ln⁡m)−A)2B/(2g(\ln m) - A)^2; for the sample median the variance is 1/(4g(ln⁡m)2)1/(4g(\ln m)^2). The ratio of these two limiting variances, the asymptotic relative efficiency of the sMAPE minimiser to the sample median, is ARE=(2g(ln⁡m)−A)24g(ln⁡m)2B;\mathrm{ARE} = \frac{(2g(\ln m) - A)^2}{4\,g(\ln m)^2\,B}; that the relative efficiency of finite test sets tends to it is not shown here, since the variances of finite test sets need not tend to those of the limits. (c) For lognormal actuals with standard deviation σ\sigma of ln⁡y\ln y: ARE→1\mathrm{ARE} \to 1 as σ→0\sigma \to 0 and AREσ3→6(ln⁡2)22π≈7.23\mathrm{ARE}\,\sigma^3 \to 6(\ln 2)^2\sqrt{2\pi} \approx 7.23 as σ→∞\sigma \to \infty. For log-Laplace actuals with scale bb of ln⁡y\ln y: ARE→1\mathrm{ARE} \to 1 as b→0b \to 0 and AREb→3\mathrm{ARE}\,b \to 3 as b→∞b \to \infty. So on both families the sMAPE minimiser is asymptotically as precise as the sample median for narrow distributions and asymptotically loses all of its precision relative to it as the actuals spread out.

SupportS 99C20 · formal · planned test · novelty: extends earlier work, by the authors

The msMAPE term sνs_\nu is the sMAPE term ss except for actuals below ν\nu. For y≥0y \ge 0, f≥0f \ge 0 and ν>0\nu > 0: (a) where y≥νy \ge \nu, sν(y,f)=s(y,f)s_\nu(y,f) = s(y,f); so when every actual is at least ν\nu, msMAPE is sMAPE and has the same target. (b) Where y<νy < \nu, sν(y,f)=|f−y|/(f+ν)s_\nu(y,f) = |f-y|/(f+\nu): for a zero actual it is f/(f+ν)f/(f+\nu), which is 0 at f=0f = 0 and rises towards 1 as ff grows, where the sMAPE term is 1 for every f>0f > 0. (c) 0≤sν≤10 \le s_\nu \le 1, with sν<1s_\nu < 1 where y<νy < \nu; and for every yy, sν(y,f)s_\nu(y,f) is continuous in ff at f=0f = 0, so the expected msMAPE term has no jump at the zero forecast, where the expected sMAPE term jumps when zero actuals have positive probability. (d) On counts with ν≤1\nu \le 1, sνs_\nu differs from ss only for zero actuals.

Supporting claims

SupportS 100C2 · formal · planned test · novelty: new, by the authors

For y>0y > 0 and f>0f > 0: (a) replacing r=y/fr = y/f by 1/r1/r leaves the sMAPE term unchanged, so a forecast kk times too high (f=kyf = ky) and one kk times too low (f=y/kf = y/k) score the same; (b) of two forecasts the same absolute distance dd from yy (0<d<y0 < d < y), the higher one scores better: s(y,y+d)=d2y+d<d2y−d=s(y,y−d).s(y, y+d) = \frac{d}{2y+d} < \frac{d}{2y-d} = s(y, y-d).

SupportS 100C5 · formal · planned test · novelty: replicates earlier work, by the authors

Let ln⁡y\ln y have a density gg symmetric about ln⁡m\ln m, g(ln⁡m+v)=g(ln⁡m−v)g(\ln m+v) = g(\ln m-v), and unimodal: g(ln⁡m+v1)≥g(ln⁡m+v2)g(\ln m+v_1) \ge g(\ln m+v_2) whenever |v1|≤|v2||v_1| \le |v_2|, strictly on a set of positive probability. Then the median mm of yy is the only sMAPE target.

SupportS 100C8 · formal · planned test · novelty: new, by the authors

For a positive random actual yy: (a) RMSE: E[(y−f)2]\mathrm E[(y-f)^2] is finite exactly when E[y2]\mathrm E[y^2] is; when E[y2]\mathrm E[y^2] is infinite but E[y]\mathrm E[y] finite, every difference between two forecasts f1f_1 and f2f_2, E[(y−f1)2−(y−f2)2]=(f1−f2)(f1+f2−2E[y]),\mathrm E[(y-f_1)^2 - (y-f_2)^2] = (f_1-f_2)(f_1+f_2-2\,\mathrm E[y]), is finite and the mean is preferred to every other forecast; when E[y]\mathrm E[y] is infinite no forecast is preferred to all others. (b) MAE: E|y−f|\mathrm E|y-f| is finite exactly when E[y]\mathrm E[y] is; otherwise differences are finite and a median is still preferred. (c) MAPE: when E[1/y]\mathrm E[1/y] is finite, the expected MAPE is finite and minimised by a median of the actuals weighted by 1/y1/y; when E[1/y]\mathrm E[1/y] is infinite, every positive forecast has infinite expected MAPE while the forecast 0 has MAPE 100, so MAPE targets 0; when yy is 0 with positive probability, MAPE is infinite or undefined for every forecast and targets nothing.

SupportS 100C11 · formal · planned test · novelty: new, by the authors

Let the actuals be μ+ε\mu + \varepsilon, where ε\varepsilon is symmetric about 0 and its density decreases away from 0 (for example, normal). If μ>0\mu > 0, every forecast with the lowest expected sMAPE is positive; if μ<0\mu < 0, every one is negative. These forecasts stay away from 0 as μ\mu approaches 0. So as μ\mu moves from slightly below 0 to slightly above, the sMAPE target jumps from a negative forecast to a positive one.

SupportS 93C13 · descriptive · planned estimate · novelty: new, by the authors

Computed from the distributions themselves, the sMAPE target on the 34 outcome distributions of plan PL1 is the median on lognormal outcomes, lies below it when ln⁡y\ln y is skewed right and above it when skewed left (quantile levels from 0.439 to 0.556 on the unimodal positive distributions), is two forecasts, one near each mode, on equal bimodal outcomes and the larger mode's on unequal ones, lies away from zero on the side of the mean on normal outcomes straddling zero, and switches to 0 at shares of zero outcomes from 0.343 to 0.468, all below one half. On test sets the forecast sMAPE ranks first spreads widely when they are small: on lognormal outcomes with σ=1\sigma = 1, its quantile level runs from 0.16 to 0.82 (5th to 95th percentile over test sets) with 6 outcomes and from 0.47 to 0.53 with 1000.

SupportS 93C14 · descriptive · planned estimate · novelty: new, by the authors

As the spread σ\sigma of ln⁡y\ln y grows from 0.1 to 10, the expected sMAPE at the sMAPE target rises towards its bound 200 (from 7.97 to 178.2 on lognormal outcomes), the target's quantile level moves from the median to that of the mode of ln⁡y\ln y (0.372 against the mode's 0.368 at σ=10\sigma = 10 on log-Gumbel outcomes skewed right), and the ranking agreement of sMAPE, for its target against a forecast half a σ\sigma away, is lower than that of MAE: at σ=4\sigma = 4 with test sets of 12 outcomes it ranks its target first in 0.61 of the test sets, MAE in 0.82.

SupportS 99C16 · descriptive · planned estimate · novelty: new, by the authors

Computed from the distributions themselves, the Suilin target differs little from the sMAPE target away from zero and on counts, and differs near zero: it is within 0.3% of the sMAPE target on the six distributions with median about 100, and the same as the sMAPE target on every Poisson counts distribution studied; on lognormal actuals with median 1 and σ=1\sigma = 1 it is 0.985 instead of 1, and the share of zeros at which it switches to the forecast 0 is 0.470 instead of sMAPE's 0.394, while at median 100 the two shares are 0.3937 and 0.3940.

SupportS 97C21 · descriptive · planned estimate · novelty: new, by the authors

The forecasts that sMAPE, msMAPE with ν=0.1\nu = 0.1, 0.5 and 1, and the Suilin loss reward were computed from the distributions themselves on the 14 intermittent demand settings. (a) On each of the 10 count distributions, the msMAPE target with ν=0.1\nu = 0.1 and the Suilin target equal the sMAPE target. With ν=0.5\nu = 0.5 the msMAPE target is 1 instead of 0 on Poisson counts with mean 0.7 and on negative binomial counts with mean 2, and 1 instead of 2 on Poisson counts with mean 1.5. With ν=1\nu = 1 it is also 1 instead of 0 on Poisson counts with mean 0.5 and on negative binomial counts with mean 1. (b) For lognormal demand sizes with zeros, the share of zeros at which the target switches to the forecast 0 is 0.394 for sMAPE, 0.419 for msMAPE with ν=0.1\nu = 0.1, 0.470 for the Suilin loss, 0.519 for msMAPE with ν=0.5\nu = 0.5 and 0.565 with ν=1\nu = 1. (c) In test sets of 12, 52 and 365 actuals, the forecast with the lowest score is 0 no more often under msMAPE than under sMAPE, and no more often under a larger ν\nu. With 52 actuals from Poisson counts with mean 0.7, it is 0 in 0.6575 of the test sets under sMAPE, 0.523 with ν=0.1\nu = 0.1, 0.1525 with ν=0.5\nu = 0.5 and 0.0265 with ν=1\nu = 1.

Evidence

E1 · quant · other · cited by C1, C2, C5, C6, C7, C8, C9, C10

Numerical checks of the formulas of the formal claims (formula checks), lab run RN8 under the formal plan that decided them; every check is within the plan's tolerances. They check the formulas on concrete numbers; the proofs are the evidence for the claims. The rows are labelled by the hypotheses of the record's first plan, whose wording kept the factor 200: H1 checks C1; H2 and H10 check C2; H5 checks C5; H7 checks C6; H8 checks C7; H12 checks C8; H13 checks C9; H14 checks C10. H3 and H9 checked C3, and H4 and H6 checked C4, as version 2 worded them; C3, C4 and C12 as now worded are checked in E31. The two results E1.R1 and E1.R2 compare a simulated variance with the large-sample variance that C19 now proves.

  • R1 simulated n Var(ln sMAPE minimiser) / large-sample value, lognormal sigma=1 = 0.9944 99% confidence interval 0.9354 to 1.0598 (hypothesis: H11, test-set size: 1000, test sets: 4000)
  • R2 simulated n Var(ln sMAPE minimiser) / large-sample value, lognormal sigma=1 = 1.022 99% confidence interval 0.9396 to 1.0982 (hypothesis: H11, test-set size: 10000, test sets: 2000)
  • R3 maximum absolute difference from the ratio form = 0.0000 (hypothesis: H1)
  • R4 maximum absolute difference between the ratio and its reciprocal = 0.0000 (hypothesis: H2)
  • R5 maximum absolute difference from 200 tanh(|z|/2) = 0.0000 (hypothesis: H3)
  • R6 violations of monotonicity, concavity or the bound 200 on 0 <= |z| <= 12 = 0 (hypothesis: H3)
  • R7 violations of 0 <= 100|z| - s <= 25|z|^3/3 = 0 (hypothesis: H9)
  • R8 maximum of |s - 100|z| + 25|z|^3/3| / |z|^5 for 0.05 < |z| < 0.5 = 0.8331 (hypothesis: H9)
  • R9 violations of s(y, y+e) < s(y, y-e) = 0 (hypothesis: H10)
  • R10 maximum absolute difference between s(y, ky) and s(y, y/k) = 0.0000 (hypothesis: H10)
  • R11 maximum relative difference between the formula for S' and a numerical derivative = 0.0000 (hypothesis: H4)
  • R12 maximum distance in ln f between the minimiser and the median = 0.0000 (hypothesis: H5)
  • R13 number of local minima, lognormal sigma=1 and log-Laplace b=1 = 1 (hypothesis: H5)
  • R14 maximum balance residual at the minimiser (pull below minus pull above) = 0.0000 (hypothesis: H6)
  • R15 settings with one local minimum where the sign of D misplaces the target = 0 (hypothesis: H6)
  • R16 maximum distance between the one-step and the exact quantile level = 0.0025 (hypothesis: H6)
  • R17 2 g(m) - E[sech^2(Z/2) tanh(|Z|/2)] for the equal-mode bimodal setting (negative: median a local maximum) = -0.3475 (hypothesis: H7)
  • R18 number of local minima of the equal-mode bimodal setting = 2 (hypothesis: H7)
  • R19 difference between the two minimal expected sMAPEs, equal-mode bimodal setting = 0.0000 (hypothesis: H7)
  • R20 maximum distance between the formula and the bisected threshold = 0.0000 (hypothesis: H8)
  • R21 opposite-sign or one-zero pairs not scoring 200 = 0 (hypothesis: H8)
  • R22 relative distance between the 1/y-weighted median and the minimiser of expected MAPE = 0.0001 (hypothesis: H12)
  • R23 difference of expected squared errors minus (f1-f2)(f1+f2-2E[y]) = 0 (hypothesis: H12)
  • R24 sMAPE terms outside [0, 200] = 0 (hypothesis: H13)
  • R25 expected sMAPE at f = 1e-6 and f = 1e10 (both near 200) = 200 (hypothesis: H13)
  • R26 the proof's bound on the error times sigma^3, 600 zeta(3) max|g''|, log-Gumbel = 309.9 (hypothesis: H14)
  • R27 limit of sigma^3 times the relative efficiency for normal log-outcomes, from the proof = 7.226 (hypothesis: H14)
E1.T1 · zero threshold by setting; hypothesis: H8 (lab runs RN8) (hypothesis: H8)
Rowsettingzero threshold
r1lognormal sigma=0.30.4687
r2lognormal sigma=0.50.4469
r3lognormal sigma=10.394
r4lognormal sigma=20.3075
E1.T2 · maximum |S - (200 - 800 ln2 h)| times sigma^3, log-Gumbel right, sigma^3 times the asymptotic relative efficiency, lognormal by sigma; hypothesis: H14 (lab runs RN8) (hypothesis: H14)
Rowsigmamaximum |S - (200 - 800 ln2 h)| times sigma^3, log-Gumbel rightsigma^3 times the asymptotic relative efficiency, lognormal
r15.0264.55.613
r210.0292.36.743
r320.0300.17.099
r440.0302.17.193

E11 · quant · other · cited by C11

Numerical checks of the formulas of C11 (formula checks), lab run RN6 under the formal plan that decided C11; every check is within the plan's tolerances. They check the formulas on concrete numbers; the proof is the evidence for the claim.

E11.T1 · sign of the global minimiser of expected sMAPE, distance of the global minimiser from 0, lowest expected sMAPE term, local minima of expected sMAPE, smallest distance of any local minimiser from 0 by mean; hypothesis: H31 (lab runs RN6) (hypothesis: H31)
Rowmeansign of the global minimiser of expected sMAPEdistance of the global minimiser from 0lowest expected sMAPE termlocal minima of expected sMAPEsmallest distance of any local minimiser from 0
r1-0.5-10.98380.522720.5849
r2-0.2-10.83710.60920.679
r3-0.05-10.7730.652520.7335
r4-0.01-10.75690.66420.7491
r50.0110.75690.66420.7491
r60.0510.7730.652520.7336
r70.210.83710.60920.679
r80.510.98380.522720.5849
E11.T2 · expected sMAPE term at a forecast near 0 by mean, forecast; hypothesis: H31 (lab runs RN6) (hypothesis: H31)
Rowmeanforecastexpected sMAPE term at a forecast near 0
r1-0.5-0.0010.9948
r2-0.50.0010.9957
r3-0.2-0.0010.9946
r4-0.20.0010.995
r5-0.05-0.0010.9947
r6-0.050.0010.9948
r7-0.01-0.0010.9947
r8-0.010.0010.9948
r90.01-0.0010.9948
r100.010.0010.9947
r110.05-0.0010.9948
r120.050.0010.9947
r130.2-0.0010.995
r140.20.0010.9946
r150.5-0.0010.9957
r160.50.0010.9948

E13 · quant · other · cited by C13

Lab runs RN1 under plan PL1. Design: Selection design (D16) on 33 outcome distributions: inverse Gaussian (6), log-skewed (12) and lognormal (2), bimodal (3), straddling zero (3), zero-inflated lognormal (8). For each: the derived values (D15) of every local minimum of expected sMAPE with its quantile level and expected score; the one-step quantile level of H6 and the pull difference D at the median; the zero threshold of H8; the quantile level of the log-mode; which moments are finite (H12); the MAPE target (H12); each measure's preferred candidate (D14), ranked by finite differences where expected scores are infinite. Candidates: quantiles at levels 0.01 to 0.99, the mean (when finite), 0, the log-mode, the MAPE minimiser (when it exists and is not 0) and every local minimiser of expected sMAPE, merged when within 1e-4 (relative). Test sets of n = 6, 8, 12, 13, 14, 18, 24, 36, 48, 100, 250, 500, 1000 and 10000 outcomes, 1000 test sets per size, seed 3 with the setting's index and n. Measures RMSE, MAE, MAPE, sMAPE on every setting. python3 claimset-code/smape-target/targets.py --seed 3 --reps 1000 (numpy 2.5.3, matplotlib 3.11.2). Analysis: Per distribution: the derived values listed in the design. Per distribution, measure and n, over test sets, of the candidate ranked first: its quantile level's mean with a 99% confidence interval and 5th, 25th, 50th, 75th and 95th percentiles; the share choosing each candidate; the share choosing the preferred candidate with its 99% Wilson interval; the number of test sets in which every candidate scores infinite. Figures: one per distribution, a panel per measure, violins of the chosen quantile levels over n with boxplots and the mean, lines at every measure's preferred candidate and at further local minima of expected sMAPE, and a note in the panel of a measure that targets nothing. These are compared with what H4 to H8, H12 and H13 predict; no test is made and no threshold is set.

inverse Gaussian shape=1000: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is sprea…
A9 inverse Gaussian shape=1000: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=1000.png).
inverse Gaussian shape=200: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread…
A10 inverse Gaussian shape=200: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=200.png).
inverse Gaussian shape=100: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread…
A11 inverse Gaussian shape=100: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=100.png).
inverse Gaussian shape=50: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread…
A12 inverse Gaussian shape=50: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=50.png).
inverse Gaussian shape=20: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread…
A13 inverse Gaussian shape=20: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=20.png).
inverse Gaussian shape=10: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread…
A14 inverse Gaussian shape=10: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=10.png).
log-skewed left sd=0.5 skew=-2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is sp…
A15 log-skewed left sd=0.5 skew=-2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=0.5_skew=-2.png).
log-skewed left sd=0.5 skew=-1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is sp…
A16 log-skewed left sd=0.5 skew=-1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=0.5_skew=-1.png).
log-skewed left sd=0.5 skew=-0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is…
A17 log-skewed left sd=0.5 skew=-0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=0.5_skew=-0.5.png).
log-skewed right sd=0.5 skew=+2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is s…
A18 log-skewed right sd=0.5 skew=+2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=0.5_skew=_2.png).
log-skewed right sd=0.5 skew=+1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is s…
A19 log-skewed right sd=0.5 skew=+1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=0.5_skew=_1.png).
log-skewed right sd=0.5 skew=+0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is…
A20 log-skewed right sd=0.5 skew=+0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=0.5_skew=_0.5.png).
lognormal sigma=0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1…
A21 lognormal sigma=0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5.png).
log-skewed left sd=1 skew=-2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spre…
A22 log-skewed left sd=1 skew=-2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=1_skew=-2.png).
log-skewed left sd=1 skew=-1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spre…
A23 log-skewed left sd=1 skew=-1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=1_skew=-1.png).
log-skewed left sd=1 skew=-0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is sp…
A24 log-skewed left sd=1 skew=-0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=1_skew=-0.5.png).
log-skewed right sd=1 skew=+2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spr…
A25 log-skewed right sd=1 skew=+2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=1_skew=_2.png).
log-skewed right sd=1 skew=+1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spr…
A26 log-skewed right sd=1 skew=+1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=1_skew=_1.png).
log-skewed right sd=1 skew=+0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is s…
A27 log-skewed right sd=1 skew=+0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=1_skew=_0.5.png).
lognormal sigma=1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 100…
A28 lognormal sigma=1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1.png).
bimodal equal modes: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1…
A29 bimodal equal modes: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/bimodal_equal_modes.png).
bimodal left mode higher: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread o…
A30 bimodal left mode higher: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/bimodal_left_mode_higher.png).
bimodal right mode higher: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread…
A31 bimodal right mode higher: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/bimodal_right_mode_higher.png).
normal straddling zero mean=0: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spr…
A32 normal straddling zero mean=0: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/normal_straddling_zero_mean=0.png).
normal straddling zero mean=0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is s…
A33 normal straddling zero mean=0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/normal_straddling_zero_mean=0.5.png).
normal straddling zero mean=1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spr…
A34 normal straddling zero mean=1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/normal_straddling_zero_mean=1.png).
lognormal sigma=0.5 with zeros p0=0.25: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks fir…
A35 lognormal sigma=0.5 with zeros p0=0.25: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.25.png).
lognormal sigma=0.5 with zeros p0=0.42: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks fir…
A36 lognormal sigma=0.5 with zeros p0=0.42: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.42.png).
lognormal sigma=0.5 with zeros p0=0.47: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks fir…
A37 lognormal sigma=0.5 with zeros p0=0.47: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.47.png).
lognormal sigma=0.5 with zeros p0=0.6: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks firs…
A38 lognormal sigma=0.5 with zeros p0=0.6: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.6.png).
lognormal sigma=1 with zeros p0=0.25: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first…
A39 lognormal sigma=1 with zeros p0=0.25: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.25.png).
lognormal sigma=1 with zeros p0=0.37: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first…
A40 lognormal sigma=1 with zeros p0=0.37: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.37.png).
lognormal sigma=1 with zeros p0=0.42: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first…
A41 lognormal sigma=1 with zeros p0=0.42: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.42.png).
lognormal sigma=1 with zeros p0=0.6: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first…
A42 lognormal sigma=1 with zeros p0=0.6: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.6.png).
E13.T1 · The sMAPE target and related values derived from each distribution: the lowest local minimum of expected sMAPE (rank 1), its quantile level and expected sMAPE, its one-step approximation, the pull from above minus the pull from below at the median, the share of zeros at which the target switches to 0, the quantile level of the log-mode, and the MAPE target
Rowsettingderived local minimum of expected sMAPE (rank: 1)quantile level of a derived local minimum of expected sMAPE (rank: 1)expected sMAPE at a derived local minimum (rank: 1)one-step quantile level of the sMAPE targetpull from above minus pull from below at the medianzero threshold of sMAPEquantile level of the log-modemedian of the outcome distributionmean of the outcome distribution
r1inverse Gaussian shape=100095.270.499924.360.4999-0.00010.46760.49895.27100
r2inverse Gaussian shape=20080.140.497848.680.4978-0.00190.43070.482280.43100
r3inverse Gaussian shape=10066.530.492961.970.4929-0.00520.40830.459767.58100
r4inverse Gaussian shape=5048.860.48275.050.482-0.01140.38450.423351.42100
r5inverse Gaussian shape=2025.90.457888.960.457-0.02240.3570.372630.11100
r6inverse Gaussian shape=1014.020.438795.750.4362-0.030.34260.347117.85100
r7log-skewed left sd=0.5 skew=-21190.521132.310.52110.01970.45611116.6109.9
r8log-skewed left sd=0.5 skew=-1110.20.51336.740.5130.01170.44940.6472108.5111.3
r9log-skewed left sd=0.5 skew=-0.5105.10.506837.960.50680.00610.44760.5681104.2112.2
r10log-skewed right sd=0.5 skew=+284.020.478932.310.4789-0.01970.4561085.78121.3
r11log-skewed right sd=0.5 skew=+190.720.48736.740.487-0.01170.44940.352892.13116.3
r12log-skewed right sd=0.5 skew=+0.595.130.493237.960.4932-0.00610.44760.431995.93114.6
r13lognormal sigma=0.51000.538.380.50.00000.44690.5100113.3
r14log-skewed left sd=1 skew=-2151.20.556257.040.55720.04560.41681135.9135.9
r15log-skewed left sd=1 skew=-1128.30.536266.420.53650.02610.40040.6472117.8146
r16log-skewed left sd=1 skew=-0.5113.90.519169.050.51920.01330.39570.5681108.7153.7
r17log-skewed right sd=1 skew=+266.140.443857.040.4428-0.04560.4168073.58
r18log-skewed right sd=1 skew=+177.960.463866.420.4635-0.02610.40040.352884.88216.5
r19log-skewed right sd=1 skew=+0.587.790.480969.050.4808-0.01330.39570.431992.03182.7
r20lognormal sigma=11000.569.940.50.00000.3940.5100164.9
r21bimodal equal modes43.490.374283.840.50.00000.36740.25100159.2
r22bimodal left mode higher39.840.406365.210.4042-0.08390.40260.32544.22122.8
r23bimodal right mode higher2510.593765.210.59580.08390.40260.675226.1195.6
r24normal straddling zero mean=0-0.7530.2257133.40.00000
r25normal straddling zero mean=0.50.98380.6858104.50.50.5
r26normal straddling zero mean=11.2870.612878.5111
r27lognormal sigma=0.5 with zeros p0=0.251000.62578.780.50.00000.44690.62580.6284.99
r28lognormal sigma=0.5 with zeros p0=0.421000.71106.30.50.00000.44690.7157.9965.72
r29lognormal sigma=0.5 with zeros p0=0.4700.471060.50.00000.44690.73545.360.06
r30lognormal sigma=0.5 with zeros p0=0.600.6800.50.00000.44690.8045.33
r31lognormal sigma=1 with zeros p0=0.251000.625102.50.50.00000.3940.62565123.7
r32lognormal sigma=1 with zeros p0=0.371000.685118.10.50.00000.3940.68544.08103.9
r33lognormal sigma=1 with zeros p0=0.4200.421160.50.00000.3940.7133.6395.63
r34lognormal sigma=1 with zeros p0=0.600.6800.50.00000.3940.8065.95
E13.T2 · The second local minimum of expected sMAPE, where there is one (rank 2)
Rowsettingderived local minimum of expected sMAPE (rank: 2)quantile level of a derived local minimum of expected sMAPE (rank: 2)expected sMAPE at a derived local minimum (rank: 2)
r1bimodal equal modes229.90.625883.84
r2normal straddling zero mean=00.7530.7743133.4
r3normal straddling zero mean=0.5-0.58490.139159.9
r4normal straddling zero mean=1-0.46490.0715179.8
r5lognormal sigma=0.5 with zeros p0=0.2500.25150
r6lognormal sigma=0.5 with zeros p0=0.4200.42116
r7lognormal sigma=0.5 with zeros p0=0.471000.735114.3
r8lognormal sigma=0.5 with zeros p0=0.61000.8135.4
r9lognormal sigma=1 with zeros p0=0.2500.25150
r10lognormal sigma=1 with zeros p0=0.3700.37126
r11lognormal sigma=1 with zeros p0=0.421000.71124.6
r12lognormal sigma=1 with zeros p0=0.61000.8148
E13.T3 · Each measure's preferred candidate (the candidate forecast with the lowest expected score): its quantile level, and the number of candidates whose expected score is infinite
Rowsettingquantile level of the preferred candidate (measure: RMSE)quantile level of the preferred candidate (measure: MAE)quantile level of the preferred candidate (measure: MAPE)quantile level of the preferred candidate (measure: sMAPE)candidates whose expected score is infinite (measure: RMSE)candidates whose expected score is infinite (measure: MAPE)
r1inverse Gaussian shape=10000.56160.49990.37860.499900
r2inverse Gaussian shape=2000.62770.50.26590.497800
r3inverse Gaussian shape=1000.66810.50.21580.492900
r4inverse Gaussian shape=500.71380.50.17730.48200
r5inverse Gaussian shape=200.77680.50.14740.457800
r6inverse Gaussian shape=100.82190.50.1360.438700
r7log-skewed left sd=0.5 skew=-20.44440.50.250.521100
r8log-skewed left sd=0.5 skew=-10.52150.50.27990.51300
r9log-skewed left sd=0.5 skew=-0.50.56030.50.29410.506800
r10log-skewed right sd=0.5 skew=+20.750.50.370.47891020
r11log-skewed right sd=0.5 skew=+10.67480.50.33880.48700
r12log-skewed right sd=0.5 skew=+0.50.63680.50.32340.493200
r13lognormal sigma=0.50.59870.50.30850.500
r14log-skewed left sd=1 skew=-20.50.500.55620101
r15log-skewed left sd=1 skew=-10.59280.50.06550.536200
r16log-skewed left sd=1 skew=-0.50.64130.50.11550.519100
r17log-skewed right sd=1 skew=+20.50.29290.44381030
r18log-skewed right sd=1 skew=+10.80340.50.23140.46381040
r19log-skewed right sd=1 skew=+0.50.74460.50.19680.480900
r20lognormal sigma=10.69150.50.15870.500
r21bimodal equal modes0.50810.50.23410.374200
r22bimodal left mode higher0.65030.50.28410.406300
r23bimodal right mode higher0.41110.50.18480.593700
r24normal straddling zero mean=00.50.50.50.77430100
r25normal straddling zero mean=0.50.50.50.30850.68580101
r26normal straddling zero mean=10.50.50.15870.61280101
r27lognormal sigma=0.5 with zeros p0=0.250.52930.50.625077
r28lognormal sigma=0.5 with zeros p0=0.420.53640.50.71059
r29lognormal sigma=0.5 with zeros p0=0.470.55160.50.47055
r30lognormal sigma=0.5 with zeros p0=0.60.62270.60.6041
r31lognormal sigma=1 with zeros p0=0.250.68810.50.625077
r32lognormal sigma=1 with zeros p0=0.370.69450.50.685065
r33lognormal sigma=1 with zeros p0=0.420.69970.50.42059
r34lognormal sigma=1 with zeros p0=0.60.73540.60.6041
E13.T4 · The candidates a measure ranks first over test sets, for four of the distributions at test-set sizes 6, 100 and 1000: the mean and the 5th, 50th and 95th percentiles of their quantile levels, the share of test sets choosing the preferred candidate, and (MAPE) the test sets in which every candidate's score is infinite; every distribution and size is in the run's CSV and its figure
Rowsettingmeasuretest-set sizemean quantile level of the chosen candidatespercentile of the chosen candidates' quantile levels (percentile: 5)percentile of the chosen candidates' quantile levels (percentile: 50)percentile of the chosen candidates' quantile levels (percentile: 95)share of test sets choosing the preferred candidatetest sets in which every candidate's score is infinite
r1lognormal sigma=1RMSE60.62870.34950.640.880.012
r2lognormal sigma=1RMSE1000.68680.610.680.770.041
r3lognormal sigma=1RMSE10000.69110.670.69150.71050.14
r4lognormal sigma=1MAE60.49870.22640.5050.75660.0222
r5lognormal sigma=1MAE1000.49870.420.50.580.0733
r6lognormal sigma=1MAE10000.50020.470.50.530.256
r7lognormal sigma=1MAPE60.2370.01950.210.58050.006
r8lognormal sigma=1MAPE1000.16140.090.15870.240.041
r9lognormal sigma=1MAPE10000.15940.130.160.190.123
r10lognormal sigma=1sMAPE60.49880.160.510.820.007
r11lognormal sigma=1sMAPE1000.49910.40.50.60.063
r12lognormal sigma=1sMAPE10000.50030.470.50.530.201
r13inverse Gaussian shape=10RMSE60.70760.40.730.940.014
r14inverse Gaussian shape=10RMSE1000.81290.740.820.870.053
r15inverse Gaussian shape=10RMSE10000.82030.80.82190.840.175
r16inverse Gaussian shape=10MAE60.490.23550.480.76770.0192
r17inverse Gaussian shape=10MAE1000.49790.4150.50.580.09
r18inverse Gaussian shape=10MAE10000.50.470.50.520.25
r19inverse Gaussian shape=10MAPE60.19910.010.170.490.008
r20inverse Gaussian shape=10MAPE1000.13970.080.1360.20.049
r21inverse Gaussian shape=10MAPE10000.13690.120.1360.160.157
r22inverse Gaussian shape=10sMAPE60.45760.13980.43870.83050.008
r23inverse Gaussian shape=10sMAPE1000.43920.320.43870.560.025
r24inverse Gaussian shape=10sMAPE10000.4390.40.43870.480.084
r25bimodal equal modesRMSE60.53550.490.50810.690.223
r26bimodal equal modesRMSE1000.51210.50810.50810.530.53
r27bimodal equal modesRMSE10000.50880.50810.50810.510.65
r28bimodal equal modesMAE60.49630.2260.49670.760.017
r29bimodal equal modesMAE1000.50120.420.5060.58520.0431
r30bimodal equal modesMAE10000.49940.470.50.530.06
r31bimodal equal modesMAPE60.2550.040.250.470.011
r32bimodal equal modesMAPE1000.23420.180.23410.290.056
r33bimodal equal modesMAPE10000.23440.210.23410.250.164
r34bimodal equal modessMAPE60.49520.140.470.840.011
r35bimodal equal modessMAPE1000.50650.310.5850.70.024
r36bimodal equal modessMAPE10000.49980.350.60.650.077
r37lognormal sigma=1 with zeros p0=0.42RMSE60.65450.440.650.890.013
r38lognormal sigma=1 with zeros p0=0.42RMSE1000.69420.620.690.770.052
r39lognormal sigma=1 with zeros p0=0.42RMSE10000.6990.680.69970.720.124
r40lognormal sigma=1 with zeros p0=0.42MAE60.53750.420.510.75020.0227
r41lognormal sigma=1 with zeros p0=0.42MAE1000.49880.420.4950.5850.0754
r42lognormal sigma=1 with zeros p0=0.42MAE10000.49950.470.50.530.248
r43lognormal sigma=1 with zeros p0=0.42MAPE60.55240.430.560.688963
r44lognormal sigma=1 with zeros p0=0.42MAPE1001000
r45lognormal sigma=1 with zeros p0=0.42MAPE10001000
r46lognormal sigma=1 with zeros p0=0.42sMAPE60.55330.420.420.890.539
r47lognormal sigma=1 with zeros p0=0.42sMAPE1000.50860.420.420.750.695
r48lognormal sigma=1 with zeros p0=0.42sMAPE10000.43460.420.420.4330.95
E13.T5 · The MAPE target derived from each distribution (a median of the actuals weighted by 1/y), where it is finite
Rowsettingderived MAPE target
r1inverse Gaussian shape=100086.51
r2inverse Gaussian shape=20053.48
r3inverse Gaussian shape=10034.69
r4inverse Gaussian shape=5019.7
r5inverse Gaussian shape=208.334
r6inverse Gaussian shape=104.21
r7log-skewed left sd=0.5 skew=-282.43
r8log-skewed left sd=0.5 skew=-179.93
r9log-skewed left sd=0.5 skew=-0.578.8
r10log-skewed right sd=0.5 skew=+276.42
r11log-skewed right sd=0.5 skew=+176.67
r12log-skewed right sd=0.5 skew=+0.577.17
r13lognormal sigma=0.577.88
r14log-skewed left sd=1 skew=-20
r15log-skewed left sd=1 skew=-118.79
r16log-skewed left sd=1 skew=-0.529.44
r17log-skewed right sd=1 skew=+252.02
r18log-skewed right sd=1 skew=+146.02
r19log-skewed right sd=1 skew=+0.542.04
r20lognormal sigma=136.79
r21bimodal equal modes36.06
r22bimodal left mode higher35.36
r23bimodal right mode higher37.44
r24normal straddling zero mean=00
r25normal straddling zero mean=0.50
r26normal straddling zero mean=10
E13.T6 · The value of each measure's preferred candidate, for the four distributions of T4
Rowsettingmeasurevalue of the preferred candidate
r1lognormal sigma=1RMSE164.9
r2lognormal sigma=1MAE100
r3lognormal sigma=1MAPE36.79
r4lognormal sigma=1sMAPE100
r5inverse Gaussian shape=10RMSE100
r6inverse Gaussian shape=10MAE17.85
r7inverse Gaussian shape=10MAPE4.21
r8inverse Gaussian shape=10sMAPE14.02
r9bimodal equal modesRMSE159.2
r10bimodal equal modesMAE100
r11bimodal equal modesMAPE36.06
r12bimodal equal modessMAPE43.49
r13lognormal sigma=1 with zeros p0=0.42RMSE95.63
r14lognormal sigma=1 with zeros p0=0.42MAE33.63
r15lognormal sigma=1 with zeros p0=0.42sMAPE0

E14 · quant · other · cited by C14

Lab runs RN2 under plan PL4. Design: Flatness design (D18) on 30 distributions (D27): lognormal, log-Gumbel right and log-Gumbel left, with sigma 0.1, 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 10. Derived (D15) from each: the quantile levels of the sMAPE target and of the log-mode; expected sMAPE at the target and the approximation of H14 there; the number of local minima; for each measure (RMSE, MAE, MAPE, sMAPE) and forecasts e^(k sigma) times its own target (k = -1, -0.5, +0.5, +1: half and one standard deviation of ln y away), the excess expected score over the target's, relative to the target's, where both are finite, and sMAPE's absolute excess; expected sMAPE at quantile levels 0.01 to 0.99. On test sets (10000 per case, seed 5 with the setting's index): the ranking agreement (D20) of each measure between its target and those four forecasts, on the lognormal family (on which every measure has a target), at n = 12, 48, 250 and 1000; the quantile level of the sMAPE minimiser of a test set at n = 48 and 1000 on every distribution. python3 claimset-code/smape-target/flatness.py --seed 5 --reps 10000 (numpy 2.5.3, matplotlib 3.11.2). Analysis: The derived values listed in the design, with 'none' and its reason where a measure targets nothing or its expected score is infinite; the ranking agreement with its 99% Wilson interval; the mean quantile level of the test sets' sMAPE minimisers with a 99% confidence interval and its 5th, 25th, 50th, 75th and 95th percentiles. Figures: the target's quantile level over sigma beside the log-mode's and the median; expected sMAPE over the quantile level of the forecast, a line per sigma; the relative excess over sigma by measure; ranking agreement over sigma by measure and n; the spread of the test sets' sMAPE minimisers over sigma. Compared with H14 (i) and (ii), and with the other measures; no test is made and no threshold is set.

The quantile level of the sMAPE target against the spread sigma of ln y, for lognormal and log-Gumbel outcomes, beside the level of the log-mode: the target mo…
A44 The quantile level of the sMAPE target against the spread sigma of ln y, for lognormal and log-Gumbel outcomes, beside the level of the log-mode: the target moves from the median to the mode of ln y as the spread grows. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_target.png).
Expected sMAPE against the quantile level of the forecast, one curve per spread sigma, for each family: the curve flattens towards 200 as sigma grows.
A45 Expected sMAPE against the quantile level of the forecast, one curve per spread sigma, for each family: the curve flattens towards 200 as sigma grows. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_expected_sMAPE.png).
Lognormal outcomes: how much worse a forecast half a sigma or one sigma away from each measure's target scores, as a share of the score at the target, against…
A46 Lognormal outcomes: how much worse a forecast half a sigma or one sigma away from each measure's target scores, as a share of the score at the target, against sigma. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_sensitivity.png).
Ranking agreement: the share of test sets that prefer each measure's target to a forecast away from it, against sigma, for several test-set sizes.
A47 Ranking agreement: the share of test sets that prefer each measure's target to a forecast away from it, against sigma, for several test-set sizes. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_agreement.png).
Where the sMAPE minimiser of a test set falls, and how widely it spreads: its quantile level over test sets against sigma, for each family.
A48 Where the sMAPE minimiser of a test set falls, and how widely it spreads: its quantile level over test sets against sigma, for each family. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_minimiser.png).
E14.T1 · The sMAPE target as the spread of ln y grows: its quantile level beside the log-mode's, the expected sMAPE there and its large-spread approximation
Rowsettingquantile level of the sMAPE targetquantile level of the log-modeexpected sMAPE at the targetlarge-spread approximation of expected sMAPE at the targetlocal minima of expected sMAPE
r1lognormal sigma=0.10.50.57.966-20121
r2log-Gumbel right sigma=0.10.49890.36799.643-17241
r3log-Gumbel left sigma=0.10.50110.63219.643-17241
r4lognormal sigma=0.250.50.519.74-684.91
r5log-Gumbel right sigma=0.250.4940.367923.66-572.71
r6log-Gumbel left sigma=0.250.5060.632123.66-572.71
r7lognormal sigma=0.50.50.538.38-242.41
r8log-Gumbel right sigma=0.50.48110.367944.85-190.41
r9log-Gumbel left sigma=0.50.51890.632144.85-190.41
r10lognormal sigma=10.50.569.94-21.221
r11log-Gumbel right sigma=10.45440.367977.991.2461
r12log-Gumbel left sigma=10.54560.632177.991.2461
r13lognormal sigma=1.50.50.593.6652.521
r14log-Gumbel right sigma=1.50.43370.3679101.266.061
r15log-Gumbel left sigma=1.50.56630.6321101.266.061
r16lognormal sigma=20.50.5111.289.391
r17log-Gumbel right sigma=20.41870.3679117.998.931
r18log-Gumbel left sigma=20.58130.6321117.998.931
r19lognormal sigma=30.50.5134.4126.31
r20log-Gumbel right sigma=30.39990.3679139.5132.31
r21log-Gumbel left sigma=30.60010.6321139.5132.31
r22lognormal sigma=40.50.5148.5144.71
r23log-Gumbel right sigma=40.38940.3679152.5149.11
r24log-Gumbel left sigma=40.61060.6321152.5149.11
r25lognormal sigma=60.50.5164.4163.11
r26log-Gumbel right sigma=60.37920.3679167.11661
r27log-Gumbel left sigma=60.62080.6321167.11661
r28lognormal sigma=100.50.5178.2177.91
r29log-Gumbel right sigma=100.37240.3679179.9179.61
r30log-Gumbel left sigma=100.62760.6321179.9179.61
E14.T2 · Ranking agreement: the share of test sets that give a measure's target a lower score than a forecast half a sigma above it, for three of the settings; every setting, offset and size is in the run's CSV and its figure
Rowsettingtest-set sizeranking agreement (measure: RMSE, offset in sigma: 0.5)ranking agreement (measure: MAE, offset in sigma: 0.5)ranking agreement (measure: MAPE, offset in sigma: 0.5)ranking agreement (measure: sMAPE, offset in sigma: 0.5)
r1lognormal sigma=0.5120.82820.77720.73960.7557
r2lognormal sigma=0.5480.96590.94210.91180.9233
r3lognormal sigma=0.52500.99980.99990.99940.9994
r4lognormal sigma=0.510001111
r5lognormal sigma=4120.98260.82260.00120.605
r6lognormal sigma=4480.97920.97050.00590.6989
r7lognormal sigma=42500.975910.03370.8823
r8lognormal sigma=410000.977110.09280.9909
E14.T3 · How much worse a forecast half a sigma or one sigma below or above the target scores, as a share of the score at the target, for lognormal outcomes
Rowsettingoffset in sigmaexcess expected score over the target, relative (measure: RMSE)excess expected score over the target, relative (measure: MAE)excess expected score over the target, relative (measure: MAPE)excess expected score over the target, relative (measure: sMAPE)excess expected sMAPE over the target (measure: sMAPE)
r1lognormal sigma=0.5-1.00.2430.30830.30830.421216.17
r2lognormal sigma=0.5-0.50.08270.09550.09550.11314.34
r3lognormal sigma=0.50.50.13310.13310.13310.11314.34
r4lognormal sigma=0.51.00.57530.5940.5940.421216.17
r5lognormal sigma=2-1.00.00690.03170.03170.227925.34
r6lognormal sigma=2-0.50.00370.01470.01470.06487.207
r7lognormal sigma=20.50.02720.05530.05530.06487.207
r8lognormal sigma=21.00.32730.43050.43040.227925.34
r9lognormal sigma=6-1.000.00000.00000.08113.32
r10lognormal sigma=6-0.500.00000.00000.02393.937
r11lognormal sigma=60.50.00000.00000.00000.02393.937
r12lognormal sigma=61.00.00000.00000.00000.08113.32
E14.T4 · Where the sMAPE minimiser of a test set falls: the mean and the 5th, 50th and 95th percentiles of its quantile level over test sets
Rowsettingtest-set sizemean quantile level of the sMAPE minimiser of a test setpercentile of the quantile level of the sMAPE minimiser of a test set (percentile: 5)percentile of the quantile level of the sMAPE minimiser of a test set (percentile: 50)percentile of the quantile level of the sMAPE minimiser of a test set (percentile: 95)
r1lognormal sigma=0.1480.50040.37640.49980.6251
r2lognormal sigma=0.110000.50.47420.49970.5263
r3log-Gumbel right sigma=0.1480.48960.36970.48930.6081
r4log-Gumbel right sigma=0.110000.49860.47240.49870.5245
r5log-Gumbel left sigma=0.1480.50910.38970.50930.6263
r6log-Gumbel left sigma=0.110000.50120.47520.50130.5275
r7lognormal sigma=0.25480.50.37620.4990.6242
r8lognormal sigma=0.2510000.49970.47290.49990.526
r9log-Gumbel right sigma=0.25480.48970.37410.48920.6057
r10log-Gumbel right sigma=0.2510000.49420.46820.49410.5206
r11log-Gumbel left sigma=0.25480.51160.39290.51160.6293
r12log-Gumbel left sigma=0.2510000.50620.48010.50640.5321
r13lognormal sigma=0.5480.50030.37760.50020.624
r14lognormal sigma=0.510000.49990.47280.49990.5274
r15log-Gumbel right sigma=0.5480.4830.36030.4820.6085
r16log-Gumbel right sigma=0.510000.48090.45290.4810.5087
r17log-Gumbel left sigma=0.5480.51750.39350.5180.6407
r18log-Gumbel left sigma=0.510000.51870.49080.51890.5467
r19lognormal sigma=1480.49960.3590.49950.6414
r20lognormal sigma=110000.50010.46810.50020.5316
r21log-Gumbel right sigma=1480.45580.31640.45450.6006
r22log-Gumbel right sigma=110000.45460.42290.45440.4874
r23log-Gumbel left sigma=1480.54380.39570.54610.684
r24log-Gumbel left sigma=110000.54580.51310.54580.5776
r25lognormal sigma=1.5480.50020.33640.49960.6636
r26lognormal sigma=1.510000.49970.46130.49950.5376
r27log-Gumbel right sigma=1.5480.43670.27550.43250.6074
r28log-Gumbel right sigma=1.510000.43410.3950.4340.4726
r29log-Gumbel left sigma=1.5480.56170.38740.56370.7247
r30log-Gumbel left sigma=1.510000.5660.52680.56660.604
r31lognormal sigma=2480.50240.31060.50190.6923
r32lognormal sigma=210000.50010.45360.50010.5468
r33log-Gumbel right sigma=2480.42540.250.41940.6192
r34log-Gumbel right sigma=210000.41910.37320.41890.4667
r35log-Gumbel left sigma=2480.57190.37630.57720.751
r36log-Gumbel left sigma=210000.58080.53330.58150.6256
r37lognormal sigma=3480.49770.27560.49920.7205
r38lognormal sigma=310000.49940.43460.49910.5664
r39log-Gumbel right sigma=3480.41540.21210.40740.6461
r40log-Gumbel right sigma=310000.40130.33850.40050.4676
r41log-Gumbel left sigma=3480.58450.35260.59170.7901
r42log-Gumbel left sigma=310000.59860.53070.59980.6617
r43lognormal sigma=4480.49850.25810.49840.7413
r44lognormal sigma=410000.49980.41340.49990.5844
r45log-Gumbel right sigma=4480.40880.18740.40030.6567
r46log-Gumbel right sigma=410000.39170.31320.38950.4769
r47log-Gumbel left sigma=4480.58870.33870.59730.8117
r48log-Gumbel left sigma=410000.60820.52070.61060.6879
r49lognormal sigma=6480.49870.23820.49760.7646
r50lognormal sigma=610000.50040.38660.50130.613
r51log-Gumbel right sigma=6480.40830.16840.39920.6722
r52log-Gumbel right sigma=610000.38390.27970.38110.4981
r53log-Gumbel left sigma=6480.58990.31670.59860.8314
r54log-Gumbel left sigma=610000.61570.50360.61870.7188
r55lognormal sigma=10480.49990.22160.50040.7798
r56lognormal sigma=1010000.50010.35940.49970.6419
r57log-Gumbel right sigma=10480.40710.1530.39560.698
r58log-Gumbel right sigma=1010000.38050.25290.37670.5195
r59log-Gumbel left sigma=10480.59150.3040.59860.8482
r60log-Gumbel left sigma=1010000.61880.48130.62270.7469

E15 · quant · other · cited by C15

Lab runs RN7 under plan PL2. Design: Efficiency design (D17) on 16 log-symmetric distributions (D26): lognormal sigma and log-Laplace b, each 0.25, 0.5, 1, 1.5, 2, 3, 4, 6. Test sets of n = 6, 8, 12, 13, 14, 18, 24, 36, 48, 100, 250, 500, 1000 and 10000 outcomes, 10000 test sets per size, seed 4 with the setting's index and n. In each test set: the sample median and the exact sMAPE minimiser (D8). From each distribution, derived (D15): the large-sample n Var of ln of each, from H11, and their ratio. python3 claimset-code/smape-target/efficiency.py --seed 4 --reps 10000 (numpy 2.5.3, matplotlib 3.11.2). Analysis: Per distribution and n, over test sets: the variances of ln(sample median) and ln(sMAPE minimiser); the relative efficiency (D9) with a 99% bootstrap percentile interval (2000 resamples of test sets); the ratio of their mean squared errors about the true median; the mean of ln(sMAPE minimiser) minus ln(median). Per distribution: the asymptotic relative efficiency. Figure: relative efficiency over n, a panel per distribution, with its band, the asymptotic value dashed and 1 dotted. Compared with H11 and H14 (iii); no test is made and no threshold is set.

lognormal outcomes: relative efficiency of the sMAPE minimiser of a test set against the sample median, against test-set size, one panel per spread, with its 9…
A50 lognormal outcomes: relative efficiency of the sMAPE minimiser of a test set against the sample median, against test-set size, one panel per spread, with its 99% band and the large-sample value as a dashed line; below 1 the sMAPE minimiser varies more than the sample median. Drawn by lab run RN7 from its own results (lab/runs/RN7/figures/efficiency_lognormal.png).
log-Laplace outcomes: relative efficiency of the sMAPE minimiser of a test set against the sample median, against test-set size, one panel per spread, with its…
A51 log-Laplace outcomes: relative efficiency of the sMAPE minimiser of a test set against the sample median, against test-set size, one panel per spread, with its 99% band and the large-sample value as a dashed line; below 1 the sMAPE minimiser varies more than the sample median. Drawn by lab run RN7 from its own results (lab/runs/RN7/figures/efficiency_log-Laplace.png).
E15.T1 · Relative efficiency of the sMAPE minimiser of a test set against the sample median, at test-set sizes 6, 12, 100, 1000 and 10000, beside its large-sample value
Rowsettingrelative efficiency (test-set size: 6)relative efficiency (test-set size: 12)relative efficiency (test-set size: 100)relative efficiency (test-set size: 1000)relative efficiency (test-set size: 10000)asymptotic relative efficiency
r1lognormal sigma=0.250.60590.70260.89960.96620.97010.9696
r2lognormal sigma=0.50.6110.70830.88020.8850.89060.8881
r3lognormal sigma=10.60670.65550.65240.66590.65650.6601
r4lognormal sigma=1.50.54840.51230.45950.46330.450.454
r5lognormal sigma=20.5010.43920.3230.30770.30430.3073
r6lognormal sigma=30.43810.36460.20580.15570.15130.1474
r7lognormal sigma=40.41480.32720.15610.0890.07890.0778
r8lognormal sigma=60.39330.30270.12290.05080.03170.0279
r9log-Laplace b=0.250.70490.82680.97690.99180.99690.9969
r10log-Laplace b=0.50.69890.8190.94480.97790.97530.9765
r11log-Laplace b=10.6670.68870.84790.87630.88530.8953
r12log-Laplace b=1.50.57660.59470.69890.77890.78040.8035
r13log-Laplace b=20.51790.50360.6120.6830.70790.7206
r14log-Laplace b=30.46720.43650.45860.52570.56280.5902
r15log-Laplace b=40.43530.37640.34960.42670.48030.4968
r16log-Laplace b=60.41670.34920.26670.28960.34410.3753
E15.T2 · Bias and variance at test-set sizes 12 and 1000, for spreads 0.25, 1 and 6: the mean of ln(sMAPE minimiser) minus ln(median), and the variances of ln(sample median) and ln(sMAPE minimiser); every setting and size is in the run's CSV
Rowsettingtest-set sizemean of ln(sMAPE minimiser) minus ln(median)variance of ln(sample median)variance of ln(sMAPE minimiser)ratio of mean squared errors about the true median
r1lognormal sigma=0.25120.00080.00720.01030.7026
r2lognormal sigma=0.2510000.00000.00010.00010.9662
r3lognormal sigma=1120.00080.11590.17680.6555
r4lognormal sigma=110000.00010.00160.00240.6659
r5lognormal sigma=6120.00214.28714.170.3027
r6lognormal sigma=61000-0.02090.05711.1260.0507
r7log-Laplace b=0.2512-0.00040.00730.00880.8268
r8log-Laplace b=0.2510000.00020.00010.00010.9918
r9log-Laplace b=1120.00140.11910.1730.6887
r10log-Laplace b=110000.00030.00110.00120.8764
r11log-Laplace b=6120.02124.26712.220.3492
r12log-Laplace b=610000.0030.03770.13030.2896

E16 · quant · other · cited by C16

Lab runs RN10 under plan PL11. Design: Exact computation, with no sampling, of the forecasts that minimise the expected Suilin term and the expected sMAPE term (claimset-code/smape-target/msmape_targets.py, run once): by Gauss-Legendre quadrature of the density of ln y for continuous distributions (with zeros added as a point mass) and by sums for counts, minimised over forecasts f >= 0 on a grid of 0 and 4000 log-spaced points refined by golden section. Settings: away from zero, lognormal sigma 0.5 and 1, inverse Gaussian shape 10 and 100 and log-skewed sd 1 with skewness -1 and +1, all at median about 100; near zero, lognormal sigma 1 at median 1 and at median 100 with shares of zeros 0, 0.2, 0.4, 0.6 and 0.8; counts, Poisson with means 0.3, 0.5, 1, 1.5, 3 and 10. For the lognormal sigma 1 at medians 1 and 100, the share of zeros at which each measure switches to the forecast 0, by bisection. Analysis: For each setting and measure: the target, its quantile level and the expected score there; for each setting, the relative difference of the Suilin target from the sMAPE target; and the two zero thresholds. Reported as tables, with no decision rule. The plan, the run's script and its results file name the Suilin loss msMAPE, as version 2 of this paper did.

The Suilin loss beside sMAPE: near zero, the two targets as a share of the median against the share of zeros, for lognormal outcomes at medians 1 and 100, with…
A53 The Suilin loss beside sMAPE: near zero, the two targets as a share of the median against the share of zeros, for lognormal outcomes at medians 1 and 100, with the zero thresholds at median 1; away from zero, the relative difference of the Suilin target from the sMAPE target for each distribution; on Poisson counts, the two targets against the mean. The figure was drawn when the paper called the Suilin loss msMAPE, and labels it so. Drawn by lab run RN11 from the recorded results of RN10 (lab/runs/RN11/figures/msmape_targets.png).
E16.T1 · The Suilin target beside the sMAPE target, with their quantile levels and the relative difference of the Suilin target from the sMAPE target
Rowsettingtarget (measure: sMAPE)target (measure: Suilin)quantile level of the target (measure: sMAPE)quantile level of the target (measure: Suilin)relative difference of the Suilin target from the sMAPE targetshare of zero outcomes
r1lognormal sigma=0.51001000.50.50.00000
r2lognormal sigma=110099.980.50.4999-0.00020
r3inverse Gaussian shape=1014.0213.980.43870.438-0.00270
r4inverse Gaussian shape=10066.5366.520.49290.4929-0.00010
r5log-skewed left sd=1 skew=-1128.3128.20.53620.5361-0.00020
r6log-skewed right sd=1 skew=+177.9677.960.46380.46380.00000
r7lognormal median=1 sigma=110.98460.50.4938-0.01540
r8lognormal median=1 sigma=1 with zeros p0=0.210.90540.60.5683-0.09460.2
r9lognormal median=1 sigma=1 with zeros p0=0.400.17020.40.4230.4
r10lognormal median=1 sigma=1 with zeros p0=0.6000.60.600.6
r11lognormal median=1 sigma=1 with zeros p0=0.8000.80.800.8
r12lognormal median=100 sigma=110099.980.50.4999-0.00020
r13lognormal median=100 sigma=1 with zeros p0=0.210099.890.60.5997-0.00110.2
r14lognormal median=100 sigma=1 with zeros p0=0.4000.40.400.4
r15lognormal median=100 sigma=1 with zeros p0=0.6000.60.600.6
r16lognormal median=100 sigma=1 with zeros p0=0.8000.80.800.8
r17Poisson mean=0.3000.74080.740800.7408
r18Poisson mean=0.5000.60650.606500.6065
r19Poisson mean=1110.73580.735800.3679
r20Poisson mean=1.5220.80880.808800.2231
r21Poisson mean=3330.64720.647200.0498
r22Poisson mean=1010100.5830.58300.0000
E16.T2 · The share of zeros at which each measure's target switches to the forecast 0, for lognormal outcomes with sigma 1
Rowsettingzero threshold (measure: sMAPE)zero threshold (measure: Suilin)
r1lognormal median=1 sigma=10.3940.47
r2lognormal median=100 sigma=10.3940.3937

E31 · quant · other · cited by C3, C4, C12, C17, C20

Numerical checks of the formulas of the claims new or reworded in version 3 (formula checks), lab run RN13 (check_v3.py, seed 8) under the formal plan that decided C3, C4, C12, C17 and C20; every check is within the plan's tolerances. They check the formulas on concrete numbers; the proofs are the evidence for the claims. The tables are labelled by the hypotheses the run's script names: H3 (T1 and T2) checks C3; H4 (T3) checks C4; H34 (T4 and T5) checks C12; H35 (T6 and T7) checks C17; H42 (T8 and T9) checks C20. The tables of H39 (T10 to T12) and H40 (T13 to T17) check the formulas of C18 and C19, whose wording was decided after later checks (E32, E33).

E31.T1 · violations of the bound on the log-MAE of the sMAPE minimiser, largest excess of the minimiser's log-MAE over the median's, minus the bound; hypothesis: H3, test_sets: 2000 (lab runs RN13) (hypothesis: H3, test_sets: 2000)
Rowviolations of the bound on the log-MAE of the sMAPE minimiserlargest excess of the minimiser's log-MAE over the median's, minus the bound
r100.0000
E31.T2 · sum of sMAPE terms, forecast at the median, actuals 1, 1, e, e^4, e^4, sum of sMAPE terms, forecast at 1, actuals 1, 1, e, e^4, e^4; hypothesis: H3 (lab runs RN13) (hypothesis: H3)
Rowsum of sMAPE terms, forecast at the median, actuals 1, 1, e, e^4, e^4sum of sMAPE terms, forecast at 1, actuals 1, 1, e, e^4, e^4
r12.7352.39
E31.T3 · largest difference of the weights below and above the target, over their sum; hypothesis: H4, distributions: 6 (lab runs RN13) (hypothesis: H4, distributions: 6)
Rowlargest difference of the weights below and above the target, over their sum
r10.0000
E31.T4 · distributions with more than one local minimum, distributions where the target is not on the side of the median that carries more weight; hypothesis: H34, distributions: 6 (lab runs RN13) (hypothesis: H34, distributions: 6)
Rowdistributions with more than one local minimumdistributions where the target is not on the side of the median that carries more weight
r100
E31.T5 · largest distance in logs between the target and the median, lognormal and log-Laplace; hypothesis: H34, distributions: 16 (lab runs RN13) (hypothesis: H34, distributions: 16)
Rowlargest distance in logs between the target and the median, lognormal and log-Laplace
r10.0000
E31.T6 · Suilin term of a zero forecast of a zero actual; hypothesis: H35 (lab runs RN13) (hypothesis: H35)
RowSuilin term of a zero forecast of a zero actual
r10
E31.T7 · largest error of the identity of (a), largest error of the identity of (b), values of the Suilin term outside [0, 1); hypothesis: H35, pairs: 200000 (lab runs RN13) (hypothesis: H35, pairs: 200000)
Rowlargest error of the identity of (a)largest error of the identity of (b)values of the Suilin term outside [0, 1)
r10.000000
E31.T8 · largest error of the identity of (a), largest error of the identity of (b), values of the msMAPE term outside [0, 1], values of the msMAPE term not below 1 for actuals below nu, pairs with a count of at least 1 where the msMAPE and sMAPE terms differ by nu; hypothesis: H42, pairs: 200000 (lab runs RN13) (hypothesis: H42, pairs: 200000)
Rownulargest error of the identity of (a)largest error of the identity of (b)values of the msMAPE term outside [0, 1]values of the msMAPE term not below 1 for actuals below nupairs with a count of at least 1 where the msMAPE and sMAPE terms differ
r10.100000
r20.500000
r31.000000
E31.T9 · msMAPE term of a forecast of 1e-9 for a zero actual by nu; hypothesis: H42 (lab runs RN13) (hypothesis: H42)
RownumsMAPE term of a forecast of 1e-9 for a zero actual
r10.10.0000
r20.50.0000
r31.00.0000
E31.T10 · signal-to-noise ratio of sMAPE over its large-spread formula, signal-to-noise ratio of the absolute error over its limit by error_distribution, c1, c2; hypothesis: H39, sigma: 200 (lab runs RN13) (hypothesis: H39, sigma: 200)
Rowerror_distributionc1c2signal-to-noise ratio of sMAPE over its large-spread formulasignal-to-noise ratio of the absolute error over its limit
r1normal0.00.50.99990.9913
r2normal0.01.010.9956
r3normal-0.51.51.0010.9966
r4Gumbel0.00.50.99980.9647
r5Gumbel0.01.010.9937
r6Gumbel-0.51.510.9967
E31.T11 · relative spread of the absolute log error's signal-to-noise ratio over sigma = 2, 10, 50 by error_distribution, c1, c2; hypothesis: H39 (lab runs RN13) (hypothesis: H39)
Rowerror_distributionc1c2relative spread of the absolute log error's signal-to-noise ratio over sigma = 2, 10, 50
r1normal0.00.50.0000
r2normal0.01.00.0000
r3normal-0.51.50.0000
r4Gumbel0.00.50.0000
r5Gumbel0.01.00.0000
r6Gumbel-0.51.50.0000
E31.T12 · signal-to-noise ratio of the squared error over its formula, lognormal, exp(sigma^2 / 2) times the squared error's ratio, lognormal, formula by c1, c2, sigma; hypothesis: H39 (lab runs RN13) (hypothesis: H39)
Rowc1c2sigmasignal-to-noise ratio of the squared error over its formula, lognormalexp(sigma^2 / 2) times the squared error's ratio, lognormal, formula
r10.00.50.51
r20.01.00.51
r3-0.51.50.51
r40.00.51.01
r50.01.01.01
r6-0.51.51.01
r70.00.52.01
r80.01.02.01
r9-0.51.52.01
r100.01.04.00.9907
r110.01.06.01
E31.T13 · distributions compared with the asymptotic values recorded by RN7; hypothesis: H40 (lab runs RN13) (hypothesis: H40)
Rowdistributions compared with the asymptotic values recorded by RN7
r116
E31.T14 · distributions where A is not below 2 g at the median, largest difference from the asymptotic values recorded by RN7; hypothesis: H40, distributions: 16 (lab runs RN13) (hypothesis: H40, distributions: 16)
Rowdistributions where A is not below 2 g at the medianlargest difference from the asymptotic values recorded by RN7
r100.0001
E31.T15 · large-sample relative efficiency, formula by family, spread; hypothesis: H40 (lab runs RN13) (hypothesis: H40)
Rowfamilyspreadlarge-sample relative efficiency, formula
r1lognormal0.250.9696
r2lognormal0.50.8881
r3lognormal1.00.6601
r4lognormal1.50.454
r5lognormal2.00.3073
r6lognormal3.00.1474
r7lognormal4.00.0778
r8lognormal6.00.0279
r9log-Laplace0.250.9969
r10log-Laplace0.50.9765
r11log-Laplace1.00.8953
r12log-Laplace1.50.8036
r13log-Laplace2.00.7207
r14log-Laplace3.00.5903
r15log-Laplace4.00.4969
r16log-Laplace6.00.3755
E31.T16 · distance of the large-sample relative efficiency from 1 at spread 0.01 by family; hypothesis: H40 (lab runs RN13) (hypothesis: H40)
Rowfamilydistance of the large-sample relative efficiency from 1 at spread 0.01
r1lognormal0.0001
r2log-Laplace0.0000
E31.T17 · large-sample relative efficiency times sigma^3 over 6 (ln 2)^2 sqrt(2 pi), lognormal, large-sample relative efficiency times b over 3, log-Laplace; hypothesis: H40, spread: 200 (lab runs RN13) (hypothesis: H40, spread: 200)
Rowlarge-sample relative efficiency times sigma^3 over 6 (ln 2)^2 sqrt(2 pi), lognormallarge-sample relative efficiency times b over 3, log-Laplace
r10.99980.9906

E32 · quant · other · cited by C18

Numerical checks of the formulas of C18 (formula checks), lab run RN18 (check_h39.py) under the formal plan that decided C18; every check is within the plan's tolerances. They check the formulas on concrete numbers; the proof is the evidence for the claim. The run's script labels the checks H39, the record's first wording of C18; the formulas are those of C18.

E32.T1 · distance of sMAPE's signal-to-noise ratio over its large-spread formula from 1, distance of the absolute error's signal-to-noise ratio over its limit from 1 by error_distribution, c1, c2, sigma; hypothesis: H39 (lab runs RN18) (hypothesis: H39)
Rowerror_distributionc1c2sigmadistance of sMAPE's signal-to-noise ratio over its large-spread formula from 1distance of the absolute error's signal-to-noise ratio over its limit from 1
r1normal0.00.52000.00010.0087
r2normal0.00.520000.00000.0009
r3normal0.01.02000.00040.0044
r4normal0.01.020000.00000.0004
r5normal-0.51.52000.00120.0034
r6normal-0.51.520000.00010.0003
r7Gumbel0.00.52000.00020.0353
r8Gumbel0.00.520000.00000.0035
r9Gumbel0.01.02000.00020.0063
r10Gumbel0.01.020000.00000.0006
r11Gumbel-0.51.52000.00040.0033
r12Gumbel-0.51.520000.00000.0003
E32.T2 · relative spread of the absolute log error's signal-to-noise ratio over sigma = 2, 10, 50 by error_distribution, c1, c2; hypothesis: H39 (lab runs RN18) (hypothesis: H39)
Rowerror_distributionc1c2relative spread of the absolute log error's signal-to-noise ratio over sigma = 2, 10, 50
r1normal0.00.50.0000
r2normal0.01.00.0000
r3normal-0.51.50.0000
r4Gumbel0.00.50.0000
r5Gumbel0.01.00.0000
r6Gumbel-0.51.50.0000
E32.T3 · distance of the squared error's signal-to-noise ratio over its formula from 1, lognormal, distance of exp(sigma^2 / 2) times the squared error's ratio from 1, lognormal, formula by c1, c2, sigma; hypothesis: H39 (lab runs RN18) (hypothesis: H39)
Rowc1c2sigmadistance of the squared error's signal-to-noise ratio over its formula from 1, lognormaldistance of exp(sigma^2 / 2) times the squared error's ratio from 1, lognormal, formula
r10.00.50.50.0000
r20.01.00.50.0000
r3-0.51.50.50.0000
r40.00.51.00.0000
r50.01.01.00.0000
r6-0.51.51.00.0000
r70.00.52.00.0000
r80.01.02.00.0000
r9-0.51.52.00.0000
r100.01.04.00.0093
r110.01.06.00.0000
r120.01.08.00.0000

E33 · quant · other · cited by C19

Numerical checks of the formulas of C19 (formula checks), lab run RN16 (check_corrected.py, seed 8) under the formal plan that decided C19; every check is within the plan's tolerances. They check the formulas on concrete numbers; the proof is the evidence for the claim. The run's script labels the checks by the record's first wordings: H40 (T4 to T7 and the result R1) checks C19; H39 (T1 to T3) checks the formulas of C18, which the same plan tested in an earlier wording.

  • R1 distributions compared with the asymptotic values recorded by RN7 = 16 (hypothesis: H40)
E33.T1 · distance of sMAPE's signal-to-noise ratio over its large-spread formula from 1, distance of the absolute error's signal-to-noise ratio over its limit from 1 by error_distribution, c1, c2, sigma; hypothesis: H39 (lab runs RN16) (hypothesis: H39)
Rowerror_distributionc1c2sigmadistance of sMAPE's signal-to-noise ratio over its large-spread formula from 1distance of the absolute error's signal-to-noise ratio over its limit from 1
r1normal0.00.52000.00010.0087
r2normal0.00.520000.00000.0009
r3normal0.01.02000.00040.0044
r4normal0.01.020000.00000.0004
r5normal-0.51.52000.00120.0034
r6normal-0.51.520000.00010.0003
r7Gumbel0.00.52000.00020.0353
r8Gumbel0.00.520000.00000.0035
r9Gumbel0.01.02000.00020.0063
r10Gumbel0.01.020000.00000.0006
r11Gumbel-0.51.52000.00040.0033
r12Gumbel-0.51.520000.00000.0003
E33.T2 · relative spread of the absolute log error's signal-to-noise ratio over sigma = 2, 10, 50 by error_distribution, c1, c2; hypothesis: H39 (lab runs RN16) (hypothesis: H39)
Rowerror_distributionc1c2relative spread of the absolute log error's signal-to-noise ratio over sigma = 2, 10, 50
r1normal0.00.50.0000
r2normal0.01.00.0000
r3normal-0.51.50.0000
r4Gumbel0.00.50.0000
r5Gumbel0.01.00.0000
r6Gumbel-0.51.50.0000
E33.T3 · distance of the squared error's signal-to-noise ratio over its formula from 1, lognormal, distance of exp(sigma^2 / 2) times the squared error's ratio from 1, lognormal, formula by c1, c2, sigma; hypothesis: H39 (lab runs RN16) (hypothesis: H39)
Rowc1c2sigmadistance of the squared error's signal-to-noise ratio over its formula from 1, lognormaldistance of exp(sigma^2 / 2) times the squared error's ratio from 1, lognormal, formula
r10.00.50.50.0000
r20.01.00.50.0000
r3-0.51.50.50.0000
r40.00.51.00.0000
r50.01.01.00.0000
r6-0.51.51.00.0000
r70.00.52.00.0000
r80.01.02.00.0000
r9-0.51.52.00.0000
r100.01.04.00.0093
r110.01.06.00.0000
r120.01.08.00.0000
E33.T4 · large-sample relative efficiency, formula by family, spread; hypothesis: H40 (lab runs RN16) (hypothesis: H40)
Rowfamilyspreadlarge-sample relative efficiency, formula
r1lognormal0.250.9696
r2lognormal0.50.8881
r3lognormal1.00.6601
r4lognormal1.50.454
r5lognormal2.00.3073
r6lognormal3.00.1474
r7lognormal4.00.0778
r8lognormal6.00.0279
r9log-Laplace0.250.9969
r10log-Laplace0.50.9765
r11log-Laplace1.00.8953
r12log-Laplace1.50.8036
r13log-Laplace2.00.7207
r14log-Laplace3.00.5903
r15log-Laplace4.00.4969
r16log-Laplace6.00.3755
E33.T5 · distance of the large-sample relative efficiency from 1 at spread 0.01 by family; hypothesis: H40 (lab runs RN16) (hypothesis: H40)
Rowfamilydistance of the large-sample relative efficiency from 1 at spread 0.01
r1lognormal0.0001
r2log-Laplace0.0000
E33.T6 · distributions where A is not below 2 g at the median, largest difference from the asymptotic values recorded by RN7; hypothesis: H40, distributions: 16 (lab runs RN16) (hypothesis: H40, distributions: 16)
Rowdistributions where A is not below 2 g at the medianlargest difference from the asymptotic values recorded by RN7
r100.0001
E33.T7 · large-sample relative efficiency times sigma^3 over 6 (ln 2)^2 sqrt(2 pi), lognormal, large-sample relative efficiency times b over 3, log-Laplace; hypothesis: H40, spread: 200 (lab runs RN16) (hypothesis: H40, spread: 200)
Rowlarge-sample relative efficiency times sigma^3 over 6 (ln 2)^2 sqrt(2 pi), lognormallarge-sample relative efficiency times b over 3, log-Laplace
r10.99980.9906

E34 · quant · other · cited by C21

Lab runs RN15 under plan PL14. Design: Which forecast does msMAPE reward on intermittent demand, beside sMAPE and the Suilin loss? One run of python3 claimset-code/smape-target/msmape_intermittent.py --seed 9, in two parts, for five measures: sMAPE, msMAPE with nu = 0.1, 0.5 and 1, and the Suilin loss, on the 14 intermittent demand settings. (1) Exact, with no sampling: the forecast that minimises the expected term of each measure (sums for counts; Gauss-Legendre quadrature of the density of ln y with the zeros as a point mass for the lognormal settings), minimised over forecasts f >= 0 on a grid of 0, the integers and 4000 log-spaced points, refined by golden section; and, for the lognormal demand sizes, the share of zeros at which each measure switches to the forecast 0, by bisection. (2) Simulated: test sets of n = 12, 52 and 365 independent actuals (a year of months, of weeks and of days), 2000 test sets per setting and n; in each test set, the forecast with the lowest value of each measure, searched over 0, the test set's own values and a grid (step 0.05 up to the largest value plus 1 for counts; 400 log-spaced points for the lognormal settings). Analysis: Part 1, per setting and measure: the target, its quantile level and the expected score there, beside the setting's share of zeros, median, mode and mean, so that the functional the target coincides with can be read; per measure, the zero threshold. Part 2, per setting, measure and n: the share of test sets whose minimiser is 0, the share whose minimiser is the exact target of part 1 (counts), and the median of the minimisers. Reported as tables and one figure (the targets against the share of zeros, a panel per measure); no test is made and no threshold is set. The figure as drawn (A56) has a panel per family of settings and a line per measure.

The forecast with the lowest expected score (the target) against the share of zero actuals, on intermittent demand: one panel per family of settings (Poisson c…
A56 The forecast with the lowest expected score (the target) against the share of zero actuals, on intermittent demand: one panel per family of settings (Poisson counts, negative binomial counts, lognormal demand sizes with zeros), one line per measure (sMAPE, msMAPE with nu = 0.1, 0.5 and 1, the Suilin loss). Drawn by lab run RN15 from its own results (lab/runs/RN15/figures/msmape_intermittent.png).
E34.T1 · The forecast each measure rewards (its target, computed from the distribution itself) on each intermittent demand setting, beside the setting's share of zeros, median, mode and mean; the measures are sMAPE, msMAPE with nu = 0.1, 0.5 and 1, and the Suilin loss
Rowsettingshare of zero outcomesmedian of the outcomesmode of the outcomesmean of the outcomestarget (measure: sMAPE)target (measure: msMAPE nu=0.1)target (measure: msMAPE nu=0.5)target (measure: msMAPE nu=1)target (measure: Suilin)
r1Poisson mean=0.10.9048000.100000
r2Poisson mean=0.30.7408000.300000
r3Poisson mean=0.50.6065000.500010
r4Poisson mean=0.70.4966100.700110
r5Poisson mean=10.367910111111
r6Poisson mean=1.50.2231111.522112
r7Poisson mean=30.049832333333
r8negative binomial mean=0.5 size=0.50.7071000.500000
r9negative binomial mean=1 size=0.50.577400100010
r10negative binomial mean=2 size=0.50.447210200110
r11lognormal sizes with zeros p0=0.20.20.72711.31910.9220.80210.86720.9054
r12lognormal sizes with zeros p0=0.40.40.38010.989200.77910.52710.59320.1702
r13lognormal sizes with zeros p0=0.60.600.659500000
r14lognormal sizes with zeros p0=0.80.800.329700000
E34.T2 · The share of zeros at which each measure's target switches to the forecast 0, for lognormal demand sizes (median 1, sigma 1) with zeros
Rowsettingzero threshold (measure: sMAPE)zero threshold (measure: msMAPE nu=0.1)zero threshold (measure: msMAPE nu=0.5)zero threshold (measure: msMAPE nu=1)zero threshold (measure: Suilin)
r1lognormal sizes with zeros0.3940.41910.51920.56470.47
E34.T3 · The quantile level of each measure's target and the expected score there (200 times the expected term), for each setting
Rowsettingquantile level of the target (measure: sMAPE)quantile level of the target (measure: msMAPE nu=0.1)quantile level of the target (measure: msMAPE nu=0.5)quantile level of the target (measure: msMAPE nu=1)quantile level of the target (measure: Suilin)expected score at the target (measure: sMAPE)expected score at the target (measure: msMAPE nu=0.1)expected score at the target (measure: msMAPE nu=0.5)expected score at the target (measure: msMAPE nu=1)expected score at the target (measure: Suilin)
r1Poisson mean=0.10.90480.90480.90480.90480.904819.0319.0319.0319.0317.34
r2Poisson mean=0.30.74080.74080.74080.74080.740851.8451.8451.8451.8447.46
r3Poisson mean=0.50.60650.60650.60650.90980.606578.6978.6978.6967.1872.37
r4Poisson mean=0.70.49660.49660.84420.84420.4966100.7100.777.8661.3192.99
r5Poisson mean=10.73580.73580.73580.73580.735894.387.6169.7857.5287.03
r6Poisson mean=1.50.80880.80880.55780.55780.808876.7774.6467.2159.7773.75
r7Poisson mean=30.64720.64720.64720.64720.647249.9149.5948.4947.4248.86
r8negative binomial mean=0.5 size=0.50.70710.70710.70710.70710.707158.5858.5858.5858.5854.46
r9negative binomial mean=1 size=0.50.57740.57740.57740.76980.577484.5384.5384.5380.3879.46
r10negative binomial mean=2 size=0.50.44720.44720.62610.62610.4472110.6110.6101.986.95105.2
r11lognormal sizes with zeros p0=0.20.60.57410.53020.55470.568395.9592.1578.3766.5389.49
r12lognormal sizes with zeros p0=0.40.40.64090.55660.58050.423120113.987.1570.994.45
r13lognormal sizes with zeros p0=0.60.60.60.60.60.68079.7872.460.9365.9
r14lognormal sizes with zeros p0=0.80.80.80.80.80.84039.8936.230.4632.95
E34.T4 · Test sets: the share of 2000 test sets of n independent actuals in which the forecast with the lowest score is 0, for each setting, n and measure
Rowsettingnshare of test sets whose minimiser is 0 (measure: sMAPE)share of test sets whose minimiser is 0 (measure: msMAPE nu=0.1)share of test sets whose minimiser is 0 (measure: msMAPE nu=0.5)share of test sets whose minimiser is 0 (measure: msMAPE nu=1)share of test sets whose minimiser is 0 (measure: Suilin)
r1Poisson mean=0.11210.99950.9960.9751
r2Poisson mean=0.15211111
r3Poisson mean=0.136511111
r4Poisson mean=0.3120.9790.9370.8310.71450.979
r5Poisson mean=0.352110.99150.91551
r6Poisson mean=0.33651110.99951
r7Poisson mean=0.5120.8480.75250.5390.3780.848
r8Poisson mean=0.5520.96850.9230.65150.2880.966
r9Poisson mean=0.5365110.84450.0671
r10Poisson mean=0.7120.60450.5150.29150.15550.6045
r11Poisson mean=0.7520.65750.5230.15250.02650.6455
r12Poisson mean=0.73650.8440.560.002500.8295
r13Poisson mean=1120.2560.22650.08950.0370.256
r14Poisson mean=1520.10150.0510.00300.1
r15Poisson mean=136500000
r16Poisson mean=1.5120.03250.030.00750.0020.0325
r17Poisson mean=1.5520.00050.0005000.0005
r18Poisson mean=1.536500000
r19Poisson mean=31200000
r20Poisson mean=35200000
r21Poisson mean=336500000
r22negative binomial mean=0.5 size=0.5120.9710.9520.8570.7210.9705
r23negative binomial mean=0.5 size=0.552110.98950.90951
r24negative binomial mean=0.5 size=0.536511111
r25negative binomial mean=1 size=0.5120.80350.780.6210.44350.808
r26negative binomial mean=1 size=0.5520.9750.96250.76750.4050.974
r27negative binomial mean=1 size=0.5365110.9750.2561
r28negative binomial mean=2 size=0.5120.5670.52650.3840.24050.5765
r29negative binomial mean=2 size=0.5520.7070.6350.30550.0680.7255
r30negative binomial mean=2 size=0.53650.9320.84750.111500.9475
r31lognormal sizes with zeros p0=0.2120.0460.0260.00550.0020.016
r32lognormal sizes with zeros p0=0.25200000
r33lognormal sizes with zeros p0=0.236500000
r34lognormal sizes with zeros p0=0.4120.46550.3740.17350.1190.303
r35lognormal sizes with zeros p0=0.4520.52350.360.0450.0120.172
r36lognormal sizes with zeros p0=0.43650.56650.207000.004
r37lognormal sizes with zeros p0=0.6120.8960.85650.6630.58050.8055
r38lognormal sizes with zeros p0=0.6520.99750.9930.8440.67650.9705
r39lognormal sizes with zeros p0=0.6365110.9990.8921
r40lognormal sizes with zeros p0=0.8120.99750.99550.97250.9590.991
r41lognormal sizes with zeros p0=0.8521110.99951
r42lognormal sizes with zeros p0=0.836511111
E34.T5 · Test sets: the share of 2000 test sets of n independent actuals in which the forecast with the lowest score is the measure's target of T1, for each count setting, n and measure
Rowsettingnshare of test sets whose minimiser is the target (measure: sMAPE)share of test sets whose minimiser is the target (measure: msMAPE nu=0.1)share of test sets whose minimiser is the target (measure: msMAPE nu=0.5)share of test sets whose minimiser is the target (measure: msMAPE nu=1)share of test sets whose minimiser is the target (measure: Suilin)
r1Poisson mean=0.11210.99950.9960.9751
r2Poisson mean=0.15211111
r3Poisson mean=0.136511111
r4Poisson mean=0.3120.9790.9370.8310.71450.979
r5Poisson mean=0.352110.99150.91551
r6Poisson mean=0.33651110.99951
r7Poisson mean=0.5120.8480.75250.5390.5470.848
r8Poisson mean=0.5520.96850.9230.65150.7120.966
r9Poisson mean=0.5365110.84450.9331
r10Poisson mean=0.7120.60450.5150.6910.80350.6045
r11Poisson mean=0.7520.65750.5230.84750.97350.6455
r12Poisson mean=0.73650.8440.560.997510.8295
r13Poisson mean=1120.53350.56650.7750.85650.537
r14Poisson mean=1520.77050.88550.98250.9930.8415
r15Poisson mean=13650.99851111
r16Poisson mean=1.5120.56750.5690.4570.5010.5685
r17Poisson mean=1.5520.73950.6570.55850.63250.6565
r18Poisson mean=1.53650.96350.8670.59650.7620.8665
r19Poisson mean=3120.5520.54750.55150.5540.548
r20Poisson mean=3520.89050.8880.8840.87950.889
r21Poisson mean=336511111
r22negative binomial mean=0.5 size=0.5120.9710.9520.8570.7210.9705
r23negative binomial mean=0.5 size=0.552110.98950.90951
r24negative binomial mean=0.5 size=0.536511111
r25negative binomial mean=1 size=0.5120.80350.780.6210.4710.808
r26negative binomial mean=1 size=0.5520.9750.96250.76750.5920.974
r27negative binomial mean=1 size=0.5365110.9750.7441
r28negative binomial mean=2 size=0.5120.5670.52650.33450.48250.5765
r29negative binomial mean=2 size=0.5520.7070.6350.4730.7820.7255
r30negative binomial mean=2 size=0.53650.9320.84750.8550.9980.9475
E34.T6 · Test sets: the median, over 2000 test sets of n independent actuals, of the forecast with the lowest score, for each setting, n and measure
Rowsettingnmedian of the test-set minimisers (measure: sMAPE)median of the test-set minimisers (measure: msMAPE nu=0.1)median of the test-set minimisers (measure: msMAPE nu=0.5)median of the test-set minimisers (measure: msMAPE nu=1)median of the test-set minimisers (measure: Suilin)
r1Poisson mean=0.11200000
r2Poisson mean=0.15200000
r3Poisson mean=0.136500000
r4Poisson mean=0.31200000
r5Poisson mean=0.35200000
r6Poisson mean=0.336500000
r7Poisson mean=0.51200010
r8Poisson mean=0.55200010
r9Poisson mean=0.536500010
r10Poisson mean=0.71200110
r11Poisson mean=0.75200110
r12Poisson mean=0.736500110
r13Poisson mean=11211111
r14Poisson mean=15211111
r15Poisson mean=136511111
r16Poisson mean=1.51222212
r17Poisson mean=1.55222112
r18Poisson mean=1.536522112
r19Poisson mean=31233333
r20Poisson mean=35233333
r21Poisson mean=336533333
r22negative binomial mean=0.5 size=0.51200000
r23negative binomial mean=0.5 size=0.55200000
r24negative binomial mean=0.5 size=0.536500000
r25negative binomial mean=1 size=0.51200010
r26negative binomial mean=1 size=0.55200010
r27negative binomial mean=1 size=0.536500010
r28negative binomial mean=2 size=0.51200110
r29negative binomial mean=2 size=0.55200110
r30negative binomial mean=2 size=0.536500110
r31lognormal sizes with zeros p0=0.2120.97910.9170.81810.87450.8357
r32lognormal sizes with zeros p0=0.2521.0030.92230.79430.86330.8825
r33lognormal sizes with zeros p0=0.23651.0030.92570.80060.86840.907
r34lognormal sizes with zeros p0=0.4120.46460.50080.51960.60030.1931
r35lognormal sizes with zeros p0=0.45200.630.52240.59130.1681
r36lognormal sizes with zeros p0=0.436500.75350.52780.59380.1719
r37lognormal sizes with zeros p0=0.61200000
r38lognormal sizes with zeros p0=0.65200000
r39lognormal sizes with zeros p0=0.636500000
r40lognormal sizes with zeros p0=0.81200000
r41lognormal sizes with zeros p0=0.85200000
r42lognormal sizes with zeros p0=0.836500000

E18 · formal · cited by C1

Proof of C1 (section H1 of the proof document A1, rendered in A2). Only the ratio of the actual to the forecast matters, not their size.

  1. Since yy and ff are positive, |y|+|f|=y+f|y| + |f| = y + f. So s=|y−f|/(y+f)s = |y - f|/(y + f). (algebra; from D1)
  2. Dividing the numerator and the denominator by f>0f > 0 gives |y−f|/f(y+f)/f\dfrac{|y-f|/f}{(y+f)/f}. Since f>0f > 0, |y−f|/f=|(y−f)/f|=|r−1||y - f|/f = |(y - f)/f| = |r - 1|. And (y+f)/f=r+1(y+f)/f = r + 1. (algebra; from S1)
  3. So s=|r−1|/(r+1)s = |r - 1|/(r + 1). The contribution depends only on the ratio of actual to forecast. (conclusion; from S2)

E19 · formal · cited by C2

Proof of C2 (section H2 of the proof document A1, rendered in A2). sMAPE is symmetric in ratios. It favours a high forecast only when errors are compared in absolute terms.

  1. By C1, s=|r−1|/(r+1)s = |r - 1|/(r + 1). Replacing rr by 1/r1/r gives |1/r−1|1/r+1\dfrac{|1/r - 1|}{1/r + 1}. Multiplying the numerator and the denominator by r>0r > 0 gives |1−r|1+r\dfrac{|1 - r|}{1 + r}, the same value. (algebra; from C1)
  2. The forecast f=kyf = ky has r=1/kr = 1/k, and f=y/kf = y/k has r=kr = k. By step 1 they score the same. This is (a). (algebra; from S1)
  3. The forecasts y+dy + d and y−dy - d are both dd away from yy, so both numerators are dd. The denominators are y+(y+d)=2y+dy + (y + d) = 2y + d and y+(y−d)=2y−d>0y + (y - d) = 2y - d > 0. The larger denominator gives the smaller value. This is (b). (algebra; from S2)
  4. So sMAPE treats a forecast kk times too high and one kk times too low alike. It favours the high forecast only when the two are equally wrong in absolute terms. (conclusion; from S3)

E20 · formal · cited by C3

Proof of C3 (section H3 of the proof document A1, rendered in A2). sMAPE measures the error on the log scale. For small errors it is half the absolute log error; for large errors it levels off.

  1. Write z=ln⁡y−ln⁡fz = \ln y - \ln f, so that r=y/f=ezr = y/f = e^z. By C1, s=|ez−1|ez+1s = \dfrac{|e^z - 1|}{e^z + 1}. (definition; from C1)
  2. Multiplying the numerator and the denominator of ez−1ez+1\dfrac{e^z - 1}{e^z + 1} by e−z/2e^{-z/2} gives ez/2−e−z/2ez/2+e−z/2=tanh⁡(z/2)\dfrac{e^{z/2} - e^{-z/2}}{e^{z/2} + e^{-z/2}} = \tanh(z/2). The denominator is positive, so s=|tanh⁡(z/2)|s = |\tanh(z/2)|. (algebra; from S1)
  3. tanh⁡\tanh changes only its sign when its argument changes sign: tanh⁡(−x)=−tanh⁡(x)\tanh(-x) = -\tanh(x), as the definition shows. So |tanh⁡(z/2)|=tanh⁡(|z|/2)|\tanh(z/2)| = \tanh(|z|/2), which is the stated form. (algebra; from S2)
  4. For (a): tanh⁡\tanh is 0 at 0. On [0,∞)[0, \infty) it is strictly increasing and concave, and it approaches 1 as its argument grows; its derivative 1−tanh2⁡x1 - \tanh^2 x is positive and decreasing for x>0x > 0. With the argument |ln⁡y−ln⁡f|/2|\ln y - \ln f|/2, ss has these properties in |ln⁡y−ln⁡f||\ln y - \ln f|. (monotonicity; from S3)
  5. For (b): for every x≥0x \ge 0, x−x3/3≤tanh⁡x≤xx - x^3/3 \le \tanh x \le x. The upper bound holds because tanh⁡\tanh has slope at most 1; the lower one because the derivative of tanh⁡x−x+x3/3\tanh x - x + x^3/3 is x2−tanh2⁡x≥0x^2 - \tanh^2 x \ge 0, and the difference is 0 at x=0x = 0. Putting x=|ln⁡y−ln⁡f|/2x = |\ln y - \ln f|/2 gives the bounds. (algebra; from S4)
  6. For (c), the median: the mean of |ln⁡yt−ln⁡f||\ln y_t - \ln f| over a test set is minimised when ln⁡f\ln f is a median of the ln⁡yt\ln y_t. Since the logarithm is increasing, this is when ff is a median mm of the yty_t. (monotonicity; from S5)
  7. For (c), the bound: with zt=ln⁡yt−ln⁡fz_t = \ln y_t - \ln f the log errors of the test set, write |z|‾(f)=1n∑t|ln⁡yt−ln⁡f|\overline{|z|}(f) = \frac1n\sum_t|\ln y_t - \ln f|, |z|3‾(f)=1n∑t|ln⁡yt−ln⁡f|3\overline{|z|^3}(f) = \frac1n\sum_t|\ln y_t - \ln f|^3 and s¯(f)\bar s(f) for the mean of the sMAPE terms of the test set. Averaging the bounds of step 5 over the test set gives s¯(f)≤12|z|‾(f)≤s¯(f)+124|z|3‾(f)\bar s(f) \le \tfrac12\overline{|z|}(f) \le \bar s(f) + \tfrac1{24}\overline{|z|^3}(f) for every f>0f > 0. Let f*f^* minimise s¯\bar s, so that s¯(f*)≤s¯(m)\bar s(f^*) \le \bar s(m). Then 12|z|‾(f*)≤s¯(f*)+124|z|3‾(f*)≤s¯(m)+124|z|3‾(f*)≤12|z|‾(m)+124|z|3‾(f*).\tfrac12\overline{|z|}(f^*) \le \bar s(f^*) + \tfrac1{24}\overline{|z|^3}(f^*) \le \bar s(m) + \tfrac1{24}\overline{|z|^3}(f^*) \le \tfrac12\overline{|z|}(m) + \tfrac1{24}\overline{|z|^3}(f^*). Multiplying by 2 gives |z|‾(f*)−|z|‾(m)≤112|z|3‾(f*)\overline{|z|}(f^*) - \overline{|z|}(m) \le \tfrac1{12}\overline{|z|^3}(f^*), the stated inequality. (algebra; from S5)
  8. For (c), not the median in general: take the five actuals 1,1,e,e4,e41, 1, e, e^4, e^4, whose median is ee. With the forecast at ee the log errors are 1,1,0,3,31, 1, 0, 3, 3 in size and the sMAPE terms sum to 2tanh⁡(0.5)+2tanh⁡(1.5)=2.7342\tanh(0.5) + 2\tanh(1.5) = 2.734. With the forecast at 1 the log errors are 0,0,1,4,40, 0, 1, 4, 4 and the terms sum to tanh⁡(0.5)+2tanh⁡(2)=2.390\tanh(0.5) + 2\tanh(2) = 2.390, which is lower. So the median does not minimise the sMAPE of this test set, although it minimises its log-MAE (step 6). (algebra; from S6)
  9. So s=tanh⁡(|ln⁡y−ln⁡f|/2)s = \tanh(|\ln y - \ln f|/2). It rises from 0 like half the absolute log error and levels off at 1. For small errors sMAPE is close to the log-MAE, and its minimiser is close to the median without in general being the median. (conclusion; from S8)

E21 · formal · cited by C9

Proof of C9 (section H4 of the proof document A1, rendered in A2). The expected values of RMSE, MAE and MAPE become infinite for every forecast when the actuals have an infinite moment, such as an infinite mean (a heavy tail, C8). The expected sMAPE never does, and there is always a best forecast.

  1. Finite. For positive values, s=tanh⁡(|ln⁡y−ln⁡f|/2)s = \tanh(|\ln y - \ln f|/2) (C3), which lies between 0 and 1. For any signs, |y−f|≤|y|+|f||y - f| \le |y| + |f|, so again s≤1s \le 1. A quantity between 0 and 1 has an expectation between 0 and 1, so the expected sMAPE is between 0 and 200, whatever the actuals. (earlier result; from C3)
  2. A best forecast. Write E[s(y,f)]\mathrm E[s(y,f)] for the expectation over the random actual yy. As ff approaches 0 or grows without bound, |ln⁡y−ln⁡f||\ln y - \ln f| grows for every positive value of yy, so ss approaches 1 (C3), and so does its expectation. For every positive ff it is below 1. It changes continuously with ff, because ss does and is at most 1 (bounded convergence). A continuous function that is below 1 in between and returns towards 1 at both ends reaches a lowest value at some f>0f > 0. A forecast of 0 or a negative forecast scores 1 on every positive actual, so it is never better. (bounded convergence; from C3, D1)
  3. So the expected sMAPE is always finite, whatever the tails of the actuals. For positive actuals, a forecast with the lowest expected sMAPE always exists. (conclusion; from S2)

E22 · formal · cited by C4

Proof of C4 (section H5 of the proof document A1, rendered in A2). For any distribution, the best forecast balances the actuals below it and above it. Each actual is weighted by how close it is to the forecast by ratio.

  1. The derivative of ss. When y>f>0y > f > 0, s=(y−f)/(y+f)s = (y - f)/(y + f). By the quotient rule its derivative in ff is −(y+f)−(y−f)(y+f)2=−2w(y,f)\dfrac{-(y+f) - (y-f)}{(y+f)^2} = -2w(y,f). When f>y>0f > y > 0, s=(f−y)/(y+f)s = (f - y)/(y + f), and in the same way its derivative is 2w(y,f)2w(y,f). Together: for y≠fy \ne f, the derivative is 2w(y,f)(𝟏{y<f}−𝟏{y>f})2w(y,f)(\mathbf 1\{y < f\} - \mathbf 1\{y > f\}). (quotient rule; from D1)
  2. A bound on it. Since (y+f)2≥4yf(y+f)^2 \ge 4yf, w(y,f)≤1/(4f)w(y,f) \le 1/(4f). So the derivative of ss is at most 1/(2f)1/(2f) in size, whatever yy, and at most 1/f1/f for every forecast above f/2f/2. The only actual at which ss has no derivative is y=fy = f, which has probability zero. (algebra; from S1)
  3. The derivative of the expectation. With this bound, the derivative may be taken inside the expectation (dominated convergence): ddfE[s(y,f)]=2E[w(y,f)(𝟏{y<f}−𝟏{y>f})].\frac{d}{df}\,\mathrm E[s(y,f)] = 2\,\mathrm E\big[w(y,f)\,(\mathbf 1\{y < f\} - \mathbf 1\{y > f\})\big]. (dominated convergence; from S2)
  4. The balance. A forecast with the lowest expected sMAPE exists (C9). There this derivative is 0, which is the stated balance: half the total weight lies below ff and half above. (earlier result; from C9)
  5. The weight. With r=y/fr = y/f, w(y,f)=1f⋅r(1+r)2=1f⋅1r+2+1/rw(y,f) = \dfrac1f\cdot\dfrac{r}{(1+r)^2} = \dfrac1f\cdot\dfrac{1}{r + 2 + 1/r}. The sum r+1/rr + 1/r is the same for rr and 1/r1/r. It is smallest, 2, at r=1r = 1, and grows as rr moves away from 1. So ww is largest at y=fy = f and falls as y/fy/f moves away from 1, equally for rr and 1/r1/r. (algebra; from S4)
  6. So sMAPE targets a weighted median of the actuals. Each actual is weighted by how close it is to the forecast by ratio. (conclusion; from S5)

E23 · formal · cited by C5

Proof of C5 (section H6 of the proof document A1, rendered in A2). This is the case of C4 in which the weights balance at the median.

  1. By C3, E[s(y,f)]=E[tanh⁡(|ln⁡y−ln⁡f|/2)]\mathrm E[s(y,f)] = \mathrm E\big[\tanh\big(|\ln y - \ln f|/2\big)\big]. Take a forecast f=meτf = m\,e^{\tau} with τ>0\tau > 0; the case τ<0\tau < 0 is the mirror image. We show that its expected sMAPE is larger than that of mm. (setup; from C3)
  2. Write both expectations as integrals over vv, the distance of ln⁡y\ln y from ln⁡f\ln f. For the forecast meτm\,e^{\tau}, ln⁡y=ln⁡m+τ+v\ln y = \ln m + \tau + v; for mm, ln⁡y=ln⁡m+v\ln y = \ln m + v. Their difference is Δ=∫tanh⁡(|v|/2)[g(ln⁡m+τ+v)−g(ln⁡m+v)]dv.\Delta = \int \tanh(|v|/2)\,\big[g(\ln m+\tau+v) - g(\ln m+v)\big]\,dv. (definition; from S1)
  3. Write ϕ(v)=g(ln⁡m+τ+v)−g(ln⁡m+v)\phi(v) = g(\ln m+\tau+v) - g(\ln m+v). By the symmetry of gg about ln⁡m\ln m, ϕ(−τ−v)=g(ln⁡m−v)−g(ln⁡m−τ−v)=g(ln⁡m+v)−g(ln⁡m+τ+v)=−ϕ(v)\phi(-\tau-v) = g(\ln m-v) - g(\ln m-\tau-v) = g(\ln m+v) - g(\ln m+\tau+v) = -\phi(v). So reflecting vv about −τ/2-\tau/2 reverses the sign of ϕ\phi. (definition; from S2)
  4. Split the integral at −τ/2-\tau/2 and reflect its left part onto the right. This gives Δ=∫v>−τ/2[tanh⁡(|v|/2)−tanh⁡(|v+τ|/2)]ϕ(v)dv.\Delta = \int_{v > -\tau/2} \big[\tanh(|v|/2) - \tanh(|v+\tau|/2)\big]\,\phi(v)\,dv. (algebra; from S3)
  5. For v>−τ/2v > -\tau/2, vv is closer to 0 than v+τv + \tau is: |v|<|v+τ||v| < |v + \tau|. So the bracket is negative, because tanh⁡\tanh is increasing (C3). And ϕ(v)≤0\phi(v) \le 0 by unimodality: g(ln⁡m+τ+v)≤g(ln⁡m+v)g(\ln m+\tau+v) \le g(\ln m+v). Their product is ≥0\ge 0, and >0> 0 on a set of positive probability. So Δ>0\Delta > 0. (monotonicity; from C3)
  6. So every forecast other than the median has a larger expected sMAPE. The median is the only forecast that minimises it. (conclusion; from S5)

E24 · formal · cited by C6

Proof of C6 (section H7 of the proof document A1, rendered in A2). When the actuals form two separate groups, the median falls in the gap between them, far from both. So sMAPE scores it worse than forecasts near either group.

  1. Write forecasts on the log scale as ln⁡f=ln⁡m+c\ln f = \ln m + c. By C3, the expected contribution is E[tanh⁡(|ln⁡y−ln⁡m−c|/2)]\mathrm E[\tanh(|\ln y - \ln m - c|/2)], a function of cc. (setup; from C3)
  2. Why the median is bad. Take the groups to be single points first. A forecast at the median is aa away, on the log scale, from every actual. So its expected ss is tanh⁡(a/2)\tanh(a/2). (case analysis; from S1)
  3. A forecast at the mode, ln⁡f=ln⁡m+a\ln f = \ln m + a, is exact for half the actuals and 2a2a away from the other half. So its expected ss is 12tanh⁡(a)\tfrac12\tanh(a). Since tanh⁡(a)=2tanh⁡(a/2)1+tanh2⁡(a/2)\tanh(a) = \dfrac{2\tanh(a/2)}{1 + \tanh^2(a/2)}, this is tanh⁡(a/2)1+tanh2⁡(a/2)<tanh⁡(a/2)\dfrac{\tanh(a/2)}{1 + \tanh^2(a/2)} < \tanh(a/2). The mode is better. (algebra; from S2)
  4. The expected ss changes continuously as the groups widen from points, because ss is bounded and continuous. So the mode is still better for narrow groups. (continuity; from S3)
  5. Why the median is a local maximum. By C12, step 4, the second derivative of the expected contribution at the median is g(ln⁡m)−12E[sech2⁡(ln⁡y−ln⁡m2)tanh⁡(|ln⁡y−ln⁡m|2)].g(\ln m) - \tfrac12\,\mathrm E\Big[\operatorname{sech}^2\Big(\frac{\ln y - \ln m}{2}\Big)\tanh\Big(\frac{|\ln y - \ln m|}{2}\Big)\Big]. (earlier result; from C12)
  6. By symmetry the first derivative at the median is 0. When the groups narrow, few actuals lie near the median, so g(ln⁡m)g(\ln m) tends to 0. Meanwhile the expectation tends to sech2⁡(a/2)tanh⁡(a/2)>0\operatorname{sech}^2(a/2)\tanh(a/2) > 0. So the second derivative becomes negative, and the median is a local maximum. (symmetry; from S5)
  7. A minimiser on each side. The expected sMAPE has a lowest point (C9). It is not at the local maximum c=0c = 0. By symmetry, if c*c^* is a minimiser, so is −c*-c^*. (symmetry; from C9)
  8. Near the modes. Write tanh⁡(|z|/2)=1−2λ(z)\tanh(|z|/2) = 1 - 2\lambda(z), with λ(z)=1/(1+e|z|)\lambda(z) = 1/(1 + e^{|z|}). The function λ\lambda is largest at 0 and falls as |z||z| grows, more slowly the larger |z||z| is. Minimising the expected ss is maximising E[λ(ln⁡y−ln⁡m−c)]\mathrm E[\lambda(\ln y - \ln m - c)]. As the groups narrow, this approaches pλ(c+a)+(1−p)λ(c−a)p\,\lambda(c + a) + (1 - p)\,\lambda(c - a), uniformly in cc. (definition; from S7)
  9. With equal weights, take 0<c<a0 < c < a. The derivative of this limit is half the rate at which λ\lambda falls at distance a−ca - c, minus half the rate at distance a+ca + c. It is positive, because λ\lambda falls faster nearer 0. For c>ac > a the limit decreases. So on c>0c > 0 its only maximum is at c=ac = a, and the minimisers on that side approach aa. By symmetry, those on the other side approach −a-a. This proves (a). (symmetry; from S8)
  10. Unequal weights. The limit increases up to c=−ac = -a and decreases after c=ac = a. In between it is convex, by the argument of step 9. So its maximum is at −a-a or at aa. The difference between these two values is (2p−1)(λ(0)−λ(2a))(2p - 1)\big(\lambda(0) - \lambda(2a)\big), which is positive when p>1/2p > 1/2. So the maximum, and with it the best forecast for narrow groups, is at the mode of the larger group. This proves (b). (convexity; from S9)
  11. So with two separate groups of actuals, the median is a hump of the expected sMAPE. The best forecasts are near the modes, and with unequal groups the larger one wins. As the weights pass through equality, the target jumps from one mode to the other. (conclusion; from S10)

E25 · formal · cited by C7

Proof of C7 (section H8 of the proof document A1, rendered in A2). The worst possible score goes to forecasts of the wrong sign and to zero forecasts. With enough zeros among the actuals, sMAPE prefers the forecast 0. The share of zeros needed is below one half.

  1. When yy and ff have opposite signs, |y−f|=|y|+|f||y - f| = |y| + |f|, so s=1s = 1. When exactly one of them is 0, the numerator and the denominator both equal the absolute value of the other one, so s=1s = 1. (algebra; from D1)
  2. A positive forecast scores 1 on every zero actual (step 1). So its expected ss is p0p_0 plus (1−p0)(1 - p_0) times its expected ss on the positive actuals. This is at least p0+(1−p0)s+p_0 + (1-p_0)\,s_+, with equality at the best positive forecasts, which exist (C9). (earlier result; from S1, C9)
  3. The forecast 0 scores 0 on zero actuals and 1 on positive ones. So its expected ss is 1−p01 - p_0. A negative forecast scores 1 on every actual, which is never better than 0. (algebra; from S2)
  4. The forecast 0 is better when 1−p0<p0+(1−p0)s+1 - p_0 < p_0 + (1 - p_0)s_+. Collecting the terms in p0p_0 gives 1−s+<p0(2−s+)1 - s_+ < p_0(2 - s_+). Since 2−s+>02 - s_+ > 0, this is the stated inequality. (algebra; from S3)
  5. Compare the threshold with 1/21/2: 1−s+2−s+<12\dfrac{1 - s_+}{2 - s_+} < \dfrac12 exactly when 2−2s+<2−s+2 - 2s_+ < 2 - s_+, that is, when s+>0s_+ > 0. And s+>0s_+ > 0 unless the positive actuals are all one value. (algebra; from S4)
  6. So sMAPE switches to the forecast 0 once zeros are frequent enough, and the share needed is below one half. At the threshold, the target jumps from a positive forecast to 0. (conclusion; from S5)

E26 · formal · cited by C11

Proof of C11 (section H9 of the proof document A1, rendered in A2). When actuals can be negative or positive, sMAPE's target is never near zero. It is on the same side of zero as the mean. So a small shift of the mean across zero makes the target jump from a negative value to a positive one.

  1. A positive forecast against its mirror image. Take f>0f > 0 and compare it with −f-f. An actual y>0y > 0 has the sign of ff, so it scores less than 1 with ff, and exactly 1 with −f-f (C7). (symmetry; from C7)
  2. The mirror actual −y-y scores 1 with ff. By symmetry, it scores the same amount less than 1 with −f-f. So each pair of actuals yy and −y-y favours ff when yy is the more likely of the two, and −f-f otherwise. (symmetry; from S1)
  3. When μ>0\mu > 0, every positive actual yy is more likely than −y-y, because yy is closer to μ\mu and the density of the actuals falls away from μ\mu. So every pair favours ff, and ff has the lower expected sMAPE. Every best forecast is therefore positive. The case μ<0\mu < 0 is the mirror image. (symmetry; from S2)
  4. Best forecasts exist on each side. For a forecast f>0f > 0, the actuals of its sign score below 1 and have positive probability, so its expected ss is below 1. As ff approaches 0 or grows without bound, every non-zero actual scores close to 1, so the expected ss approaches 1. As in C9, the expected ss is continuous in ff and so reaches a lowest value for f>0f > 0; the same holds for f<0f < 0. (continuity; from C9)
  5. Away from zero, uniformly. Fix a forecast f1>0f_1 > 0. For every μ\mu in a range around 0, say |μ|≤1|\mu| \le 1, its expected ss is at most some value q<1q < 1, since it changes continuously with μ\mu. For a forecast ff with |f|≤η|f| \le \eta, every actual with |y|≥η|y| \ge \sqrt\eta scores at least (η−η)/(η+η)(\sqrt\eta - \eta)/(\sqrt\eta + \eta), which approaches 1 as η\eta shrinks; and the share of actuals with |y|<η|y| < \sqrt\eta is small for every such μ\mu: since ε\varepsilon is symmetric and its density decreases away from 0, an interval of a given length around 0 holds at least as much of its probability as any other interval of that length, so P(|μ+ε|<η)≤P(|ε|<η)P(|\mu + \varepsilon| < \sqrt\eta) \le P(|\varepsilon| < \sqrt\eta), which tends to 0 as η\eta shrinks. So for η\eta small enough, every forecast within η\eta of 0 has an expected ss above qq, for every μ\mu in the range: it is worse than f1f_1 (or its mirror image). The best forecasts therefore stay at least η\eta from 0. (symmetry; from S4)
  6. So the target is always on the same side of zero as the mean, and never near zero. It jumps from negative to positive as the mean crosses zero. (conclusion; from S5)

E27 · formal · cited by C8

Proof of C8 (section H10 of the proof document A1, rendered in A2). Each of the other three measures has an infinite expected value, for every forecast, when a moment it needs is infinite. What it then rewards depends on the measure.

  1. (a) RMSE. If E[y2]\mathrm E[y^2] is finite, so is E[y]\mathrm E[y]. Expanding the square gives E[(y−f)2]=E[y2]−2fE[y]+f2\mathrm E[(y-f)^2] = \mathrm E[y^2] - 2f\,\mathrm E[y] + f^2, which is finite. Conversely, y2≤2(y−f)2+2f2y^2 \le 2(y - f)^2 + 2f^2. So a finite E[(y−f)2]\mathrm E[(y-f)^2] makes E[y2]\mathrm E[y^2] finite. (algebra; from D10)
  2. Expanding both squares gives (y−f1)2−(y−f2)2=(f2−f1)(2y−f1−f2)(y - f_1)^2 - (y - f_2)^2 = (f_2 - f_1)(2y - f_1 - f_2). Its expectation is finite when E[y]\mathrm E[y] is, and equals (f1−f2)(f1+f2−2E[y])=(f1−E[y])2−(f2−E[y])2(f_1 - f_2)(f_1 + f_2 - 2\,\mathrm E[y]) = (f_1 - \mathrm E[y])^2 - (f_2 - \mathrm E[y])^2. (algebra; from S1)
  3. For f1=E[y]f_1 = \mathrm E[y] this is negative for every f2≠E[y]f_2 \ne \mathrm E[y]. So the mean is preferred. (algebra; from S2)
  4. If E[y]\mathrm E[y] is infinite, then for f2>f1f_2 > f_1 the expectation in step 2 is +∞+\infty. Every forecast is beaten by any larger one, so none is preferred to all others. (algebra; from S2)
  5. (b) MAE. Since |y−f|≥y−f|y - f| \ge y - f, E|y−f|\mathrm E|y - f| is infinite when E[y]\mathrm E[y] is. Otherwise it is at most E[y]+|f|\mathrm E[y] + |f|. (algebra; from D11)
  6. The difference |y−f1|−|y−f2||y - f_1| - |y - f_2| is at most |f1−f2||f_1 - f_2| in size, so its expectation is finite. (algebra; from S5)
  7. The derivative of |y−f||y - f| in ff is 𝟏{y<f}−𝟏{y>f}\mathbf 1\{y < f\} - \mathbf 1\{y > f\}. The expectation of this is P(y<f)−P(y>f)P(y < f) - P(y > f), which is negative below a median and positive above it. So a median is preferred. (algebra; from S6)
  8. (c) MAPE. For f>0f > 0, the contribution of an actual to MAPE, divided by 100, is |y−f|/y=|1−f/y||y - f|/y = |1 - f/y|. This is at least f/y−1f/y - 1. So the expected MAPE is infinite when E[1/y]\mathrm E[1/y] is, and at most 1+fE[1/y]1 + f\,\mathrm E[1/y] otherwise. (algebra; from D12)
  9. As a function of ff, |1−f/y||1 - f/y| is convex. Its derivative is (1/y)(𝟏{y<f}−𝟏{y>f})(1/y)(\mathbf 1\{y < f\} - \mathbf 1\{y > f\}). Its expectation is 0 where the actuals below ff and those above it have equal total weight 1/y1/y: at a median of the actuals weighted by 1/y1/y. (convexity; from S8)
  10. The forecast 0 gives |1−0|=1|1 - 0| = 1 for every actual, so MAPE 100. A negative forecast gives more than 1. So when E[1/y]\mathrm E[1/y] is infinite, 0 is the only forecast with a finite expected MAPE. (algebra; from S9)
  11. If y=0y = 0 with positive probability: for f≠0f \ne 0 that actual contributes |f|/0|f|/0, which is infinite. For f=0f = 0 it contributes 0/00/0, which is undefined. (algebra; from S10)
  12. So each of RMSE, MAE and MAPE has an infinite expected value when a moment it needs is infinite. RMSE and MAE still rank forecasts by their differences. MAPE turns to the forecast 0, and with zero actuals it targets nothing. (conclusion; from S11)

E28 · formal · cited by C10

Proof of C10 (section H11 of the proof document A1, rendered in A2). For widely spread actuals, every forecast scores close to the worst value. The differences between forecasts shrink, and the target moves from the median to the mode of ln⁡y\ln y.

  1. Rewrite the expectation. By C3, and since tanh⁡(|x|/2)=1−21+e|x|\tanh(|x|/2) = 1 - \dfrac{2}{1 + e^{|x|}}, E[s(y,f)]=1−E[2/(1+e|ln⁡y−ln⁡f|)]\mathrm E[s(y,f)] = 1 - \mathrm E\big[2/(1 + e^{|\ln y - \ln f|})\big]. The density of ln⁡y\ln y at xx is h((x−μ)/σ)/σh\big((x - \mu)/\sigma\big)/\sigma. Substituting x=ln⁡f+vx = \ln f + v in the integral gives E[s(y,f)]=1−1σ∫21+e|v|h(c+vσ)dv.\mathrm E[s(y,f)] = 1 - \frac1\sigma\int \frac{2}{1 + e^{|v|}}\,h\Big(c + \frac v\sigma\Big)\,dv. (change of variables; from C3)
  2. Expand hh. By Taylor's theorem, h(c+v/σ)=h(c)+h′(c)v/σ+12h′′(ξ)v2/σ2h(c + v/\sigma) = h(c) + h'(c)\,v/\sigma + \tfrac12 h''(\xi)\,v^2/\sigma^2 for some ξ\xi between cc and c+v/σc + v/\sigma. (Taylor's theorem; from S1)
  3. The first term integrates to 4ln⁡2h(c)4\ln 2\,h(c), since ∫21+e|v|dv=4ln⁡2\int \frac{2}{1+e^{|v|}}\,dv = 4\ln 2. Over v>0v > 0, ∫2/(1+ev)dv=2ln⁡2\int 2/(1 + e^v)\,dv = 2\ln 2. (algebra; from S2)
  4. The second term integrates to 0: its factor vv changes sign with vv, while the rest does not. (algebra; from S3)
  5. The third term is at most 12max⁡|h′′|\tfrac12\max|h''| times ∫2v21+e|v|dv=6ζ(3)≈7.21\int \frac{2v^2}{1+e^{|v|}}\,dv = 6\zeta(3) \approx 7.21, divided by σ2\sigma^2. This integral is a standard Fermi–Dirac integral. With the factor 1/σ1/\sigma of step 1, this gives the expansion. (standard integral; from S1)
  6. (a). The expansion tends to 1, so the expected sMAPE tends to 200. Two forecasts at fixed quantile levels have fixed positions c1c_1 and c2c_2. So their difference is 4ln⁡2[h(c2)−h(c1)]/σ+O(σ−3)4\ln 2\,[h(c_2) - h(c_1)]/\sigma + O(\sigma^{-3}). (algebra; from S5)
  7. (b). Minimising the expansion over cc is maximising h(c)h(c), up to a term that shrinks like σ−2\sigma^{-2} for every cc at once. A density with a bounded derivative tends to 0 far out. So its single highest point c0c_0 stands clear of every other point, and the minimisers' positions tend to c0c_0. So do their quantile levels. (algebra; from S6)
  8. (c). By C3 (b), 12σ|ε−c|−s\tfrac12\sigma|\varepsilon - c| - s lies between 0 and 124σ3|ε−c|3\tfrac1{24}\sigma^3|\varepsilon - c|^3. So for cc in any bounded range, E[s]/σ=12E|ε−c|+O(σ2)\mathrm E[s]/\sigma = \tfrac12\mathrm E|\varepsilon - c| + O(\sigma^2). This uses E|ε|3<∞\mathrm E|\varepsilon|^3 < \infty. And E|ε−c|\mathrm E|\varepsilon - c| is smallest at the median of ε\varepsilon. (earlier result; from C3)
  9. The minimisers also stay in a bounded range. For large |c||c|, most actuals have ε\varepsilon at least |c|/2|c|/2 from cc. And tanh⁡x≥xtanh⁡1\tanh x \ge x\tanh 1 for 0≤x≤10 \le x \le 1. So E[s]/σ\mathrm E[s]/\sigma grows with |c||c|, or exceeds a multiple of 1/σ1/\sigma, beyond its value at the median. So the minimisers' positions tend to the median of ε\varepsilon, whose quantile level is 1/21/2. (algebra; from S8)
  10. So as the actuals spread out, the expected sMAPE flattens towards its bound. Differences between forecasts vanish like 1/σ1/\sigma. The target moves from the median to the mode of ln⁡y\ln y. (conclusion; from S9)

E29 · formal · cited by C12

Proof of C12 (section H12 of the proof document A1, rendered in A2). The target is above the ordinary median when the actuals above it carry more weight, and below it when those below carry more. When the weights balance and enough actuals lie near the median, the median is the target. This is so when the log of the actual is symmetric with a single peak.

  1. The derivative at the median. By C4, step 3, the derivative of the expected sMAPE term in ff is 2E[w(y,f)(𝟏{y<f}−𝟏{y>f})]2\,\mathrm E[w(y,f)(\mathbf 1\{y < f\} - \mathbf 1\{y > f\})]. At f=mf = m it is twice the total weight of the actuals below mm minus that of the actuals above mm. (earlier result; from C4)
  2. The shape of expected sMAPE. With a single local minimum, the expected sMAPE decreases up to it and increases after it. Otherwise it would turn again, at a local maximum. After that turn it would have to fall and then rise once more, because it returns towards 1 at both ends (C9). That would give a second local minimum. (earlier result; from C9)
  3. (a). If the actuals above mm carry more weight, the derivative at the median is negative (step 1): the expected sMAPE is decreasing at the median. By step 2, the median then lies before the minimum, so the target is above the median. The case where the actuals below carry more weight is the mirror image. (symmetry; from S1, S2)
  4. The second derivative. Measure the forecast on the log scale, c=ln⁡fc = \ln f, so that the expected contribution is E[tanh⁡(|ln⁡y−c|/2)]\mathrm E[\tanh(|\ln y - c|/2)] (C3). The derivative of tanh⁡(|x−c|/2)\tanh(|x - c|/2) in cc is −12sech2⁡(x−c2)-\tfrac12\operatorname{sech}^2\big(\tfrac{x-c}{2}\big) for c<xc < x and +12sech2⁡(x−c2)+\tfrac12\operatorname{sech}^2\big(\tfrac{x-c}{2}\big) for c>xc > x: it jumps by 1 at c=xc = x. Differentiating again and taking the expectation over x=ln⁡yx = \ln y, the jumps add the density gg of ln⁡y\ln y at cc, and elsewhere the second derivative is −12sech2⁡(x−c2)tanh⁡(|x−c|2)-\tfrac12\operatorname{sech}^2\big(\tfrac{x-c}{2}\big)\tanh\big(\tfrac{|x-c|}{2}\big). So the second derivative of the expected contribution is g(c)−12E[sech2⁡(ln⁡y−c2)tanh⁡(|ln⁡y−c|2)].g(c) - \tfrac12\,\mathrm E\Big[\operatorname{sech}^2\Big(\frac{\ln y - c}{2}\Big)\tanh\Big(\frac{|\ln y - c|}{2}\Big)\Big]. This holds when gg is continuous at cc: since sech2⁡=1−tanh2⁡\operatorname{sech}^2 = 1 - \tanh^2, the first derivative is P(ln⁡y≤c)−12\mathrm P(\ln y \le c) - \tfrac12, whose derivative at cc is g(c)g(c), plus the expectation of −12tanh2⁡(x−c2)-\tfrac12\tanh^2\big(\tfrac{x-c}{2}\big) with the sign of c−xc - x, a function of cc with a bounded continuous derivative, which may be differentiated inside the expectation. (continuity; from C3)
  5. (b), the sufficient condition. If the two weights are equal, the first derivative at the median is 0 (step 1): the expected sMAPE is flat there. If, in addition, the stated inequality holds, the second derivative at c=ln⁡mc = \ln m is positive (step 4), so the median is a strict local minimum. With a single local minimum, it is the target. (algebra; from S1, S4)
  6. (b), without the inequality. A point where the first derivative is 0 can be a flat point of inflection, through which the expected sMAPE keeps rising (or keeps falling) without turning. An example, built by the independent checker of this section: let ln⁡y−ln⁡m\ln y - \ln m be a mixture of three normal distributions, with mean −1.5-1.5 and standard deviation 0.30.3 (weight 0.406125), mean 0 and standard deviation 0.50.5 (weight 0.054825), and mean 1.1586641.158664 and standard deviation 1 (weight 0.539050); these values are rounded from the exact solution of the three conditions. Its median is mm, the two weights are equal there, and the second derivative of step 4 is 0 at ln⁡m\ln m; the expected sMAPE rises through mm, where it is flat, and its only minimum is near ln⁡f=ln⁡m−1.25\ln f = \ln m - 1.25. So equal weights alone do not make the median the target. (algebra; from S4)
  7. (c). By C5 the median is then the only forecast with the lowest expected sMAPE (a density cannot be constant, so a symmetric unimodal density is strictly lower somewhere away from its centre, as C5 asks). For lognormal actuals ln⁡y\ln y has a normal density, and for log-Laplace actuals the Laplace density e−|x−ln⁡m|/b/(2b)e^{-|x - \ln m|/b}/(2b); both are symmetric about ln⁡m\ln m and strictly decreasing in the distance from it. (earlier result; from C5)
  8. So the target is above the ordinary median when the actuals above it carry more of the weight, and below it when those below carry more. When the two balance, the median is the target if, in addition, enough actuals lie near it, as the stated inequality says; this is the case when the log of the actual is symmetric with a single peak. (conclusion; from S7)

E35 · formal · cited by C18

Proof of C18 (section H13 of the proof document A1, rendered in A2). For widely spread actuals, sMAPE loses its power to tell two forecasts apart, at the rate σ−1/2\sigma^{-1/2}. The absolute log error keeps its power, and the absolute error keeps it unless the higher forecast is the median; the squared error on lognormal actuals loses it faster than any power of σ\sigma.

  1. The sMAPE difference. Write q(x)=2/(1+e|x|)q(x) = 2/(1 + e^{|x|}), so that tanh⁡(|x|/2)=1−q(x)\tanh(|x|/2) = 1 - q(x), and qi=q(σ(ε−ci))q_i = q(\sigma(\varepsilon - c_i)). By C3 the sMAPE term of the forecast fif_i is 1−qi1 - q_i, because ln⁡y−ln⁡fi=σ(ε−ci)\ln y - \ln f_i = \sigma(\varepsilon - c_i). So D=q1−q2D = q_1 - q_2. (definition; from C3)
  2. Its mean. By C10, E[qi]=4ln⁡2h(ci)/σ+O(σ−3)\mathrm E[q_i] = 4\ln 2\,h(c_i)/\sigma + O(\sigma^{-3}). So E[D]=4ln⁡2(h(c1)−h(c2))/σ+O(σ−3)\mathrm E[D] = 4\ln 2\,(h(c_1) - h(c_2))/\sigma + O(\sigma^{-3}). (earlier result; from C10)
  3. The mean of qi2q_i^2. Substituting v=σ(ε−ci)v = \sigma(\varepsilon - c_i), as in C10, step 1, E[qi2]=1σ∫q(v)2h(ci+v/σ)dv\mathrm E[q_i^2] = \frac1\sigma\int q(v)^2\,h(c_i + v/\sigma)\,dv. Expanding hh by Taylor's theorem, as in C10, step 2: the first term integrates to h(ci)∫q(v)2dvh(c_i)\int q(v)^2\,dv; the second integrates to 0, because its factor vv changes sign with vv while the rest does not; the third is at most 12max⁡|h′′|∫v2q(v)2dv\tfrac12\max|h''|\int v^2 q(v)^2\,dv divided by σ2\sigma^2, and this integral is finite because q(v)≤2e−|v|q(v) \le 2e^{-|v|}. Since 1(1+ev)2=11+ev−ev(1+ev)2\dfrac{1}{(1+e^v)^2} = \dfrac{1}{1+e^v} - \dfrac{e^v}{(1+e^v)^2}, whose two parts integrate over v>0v > 0 to ln⁡2\ln 2 and 12\tfrac12, ∫q(v)2dv=8(ln⁡2−12)=4(2ln⁡2−1)\int q(v)^2\,dv = 8(\ln 2 - \tfrac12) = 4(2\ln 2 - 1). So E[qi2]=4(2ln⁡2−1)h(ci)/σ+O(σ−3)\mathrm E[q_i^2] = 4(2\ln 2 - 1)\,h(c_i)/\sigma + O(\sigma^{-3}). (Taylor's theorem; from C10)
  4. The cross term. Since q(x)≤2e−|x|q(x) \le 2e^{-|x|}, q1q2≤4e−σ(|ε−c1|+|ε−c2|)≤4e−σ(c2−c1)q_1 q_2 \le 4e^{-\sigma(|\varepsilon - c_1| + |\varepsilon - c_2|)} \le 4e^{-\sigma(c_2 - c_1)}, because |ε−c1|+|ε−c2|≥c2−c1|\varepsilon - c_1| + |\varepsilon - c_2| \ge c_2 - c_1. So E[q1q2]≤4e−σ(c2−c1)\mathrm E[q_1 q_2] \le 4e^{-\sigma(c_2 - c_1)}, which is smaller than every power of 1/σ1/\sigma. (algebra; from S3)
  5. The variance. Var⁡(D)=E[q12]+E[q22]−2E[q1q2]−(E[D])2\operatorname{Var}(D) = \mathrm E[q_1^2] + \mathrm E[q_2^2] - 2\,\mathrm E[q_1 q_2] - (\mathrm E[D])^2. By steps 2 to 4 this is 4(2ln⁡2−1)(h(c1)+h(c2))/σ+O(σ−2)4(2\ln 2 - 1)(h(c_1) + h(c_2))/\sigma + O(\sigma^{-2}); the term (E[D])2(\mathrm E[D])^2 is of order σ−2\sigma^{-2}. (algebra; from S2, S3, S4)
  6. (a). When h(c1)≠h(c2)h(c_1) \ne h(c_2), h(c1)+h(c2)>0h(c_1) + h(c_2) > 0. Dividing the size of the mean (step 2) by the square root of the variance (step 5) gives the stated ratio. To keep n\sqrt n times the ratio at a given value, nn must be that value squared over the ratio squared, which grows in proportion to σ\sigma. When h(c1)=h(c2)>0h(c_1) = h(c_2) > 0 instead, the mean is O(σ−3)O(\sigma^{-3}) (step 2) while the variance is still of order 1/σ1/\sigma (step 5), so the ratio is O(σ−5/2)O(\sigma^{-5/2}): at most of that order. (algebra; from S2, S5)
  7. (b). For the absolute log error, |ln⁡y−ln⁡fi|=σ|ε−ci||\ln y - \ln f_i| = \sigma|\varepsilon - c_i|, so D=σ(|ε−c2|−|ε−c1|)D = \sigma(|\varepsilon - c_2| - |\varepsilon - c_1|). It is bounded by σ(c2−c1)\sigma(c_2 - c_1), so its mean and standard deviation exist. Both are σ\sigma times those of |ε−c2|−|ε−c1||\varepsilon - c_2| - |\varepsilon - c_1|, whose distribution does not depend on σ\sigma. So their ratio, wherever the standard deviation is not zero, does not depend on σ\sigma. (algebra; from S6)
  8. (c). For the absolute error, D=|y−f2|−|y−f1|D = |y - f_2| - |y - f_1| with f1<f2f_1 < f_2. It equals f2−f1f_2 - f_1 when y≤f1y \le f_1, f1+f2−2yf_1 + f_2 - 2y when f1<y<f2f_1 < y < f_2, and −(f2−f1)-(f_2 - f_1) when y≥f2y \ge f_2; so |D|≤f2|D| \le f_2. Now f1/f2=e−σ(c2−c1)f_1/f_2 = e^{-\sigma(c_2 - c_1)} tends to 0, and for every ε<c2\varepsilon < c_2, y/f2=eσ(ε−c2)y/f_2 = e^{\sigma(\varepsilon - c_2)} tends to 0. So D/f2D/f_2 tends to 1 for ε<c2\varepsilon < c_2 and to −1-1 for ε>c2\varepsilon > c_2. By bounded convergence, E[D/f2]→H(c2)−(1−H(c2))=2H(c2)−1\mathrm E[D/f_2] \to H(c_2) - (1 - H(c_2)) = 2H(c_2) - 1 and E[(D/f2)2]→1\mathrm E[(D/f_2)^2] \to 1. So the variance of D/f2D/f_2 tends to 1−(2H(c2)−1)2=4H(c2)(1−H(c2))1 - (2H(c_2) - 1)^2 = 4H(c_2)(1 - H(c_2)), and the ratio, which is the same for DD and D/f2D/f_2, tends to the stated limit when 0<H(c2)<10 < H(c_2) < 1. The limit is 0 only when H(c2)=1/2H(c_2) = 1/2. (bounded convergence; from S7)
  9. (d). For the squared error, D=(y−f2)2−(y−f1)2=(f2−f1)(f1+f2−2y)D = (y - f_2)^2 - (y - f_1)^2 = (f_2 - f_1)(f_1 + f_2 - 2y). So E[D]=(f2−f1)(f1+f2−2E[y])\mathrm E[D] = (f_2 - f_1)(f_1 + f_2 - 2\,\mathrm E[y]) and sd(D)=2(f2−f1)sd(y)\mathrm{sd}(D) = 2(f_2 - f_1)\,\mathrm{sd}(y), finite when E[y2]\mathrm E[y^2] is. Their ratio is the stated one. For lognormal actuals E[y]=eμ+σ2/2\mathrm E[y] = e^{\mu + \sigma^2/2}, sd(y)=E[y]eσ2−1\mathrm{sd}(y) = \mathrm E[y]\sqrt{e^{\sigma^2} - 1} and fi/E[y]=eσci−σ2/2f_i/\mathrm E[y] = e^{\sigma c_i - \sigma^2/2}, which tends to 0. So eσ2/2e^{\sigma^2/2} times the ratio is |1−(f1+f2)/(2E[y])|eσ2/2/eσ2−1|1 - (f_1 + f_2)/(2\,\mathrm E[y])|\,e^{\sigma^2/2}/\sqrt{e^{\sigma^2} - 1}, which tends to 1. (algebra; from S8)
  10. So for widely spread actuals the absolute log error keeps its power to tell two forecasts apart wherever its ratio is defined. The absolute error keeps it when 0<H(c2)<10 < H(c_2) < 1 and the higher forecast is not the median. sMAPE loses it at the rate σ−1/2\sigma^{-1/2} when h(c1)≠h(c2)h(c_1) \ne h(c_2). It loses it faster, with a ratio of at most order σ−5/2\sigma^{-5/2}, when h(c1)=h(c2)>0h(c_1) = h(c_2) > 0. The squared error on lognormal actuals loses it faster than any power of σ\sigma. (conclusion; from S9)

E36 · formal · cited by C19

Proof of C19 (section H14 of the proof document A1, rendered in A2). Where the log of the actual is symmetric with a single peak, the forecast that minimises the sMAPE of a test set is asymptotically normal about the median. Compared with the sample median it is asymptotically as precise for narrow distributions and asymptotically loses all of its precision as the actuals spread out.

  1. Write xt=ln⁡yt−ln⁡mx_t = \ln y_t - \ln m for the actuals of a test set and θ=ln⁡f−ln⁡m\theta = \ln f - \ln m for a forecast ff. By C3 the mean sMAPE term of the test set is Mn(θ)=1n∑ttanh⁡(|xt−θ|/2)M_n(\theta) = \frac1n\sum_t\tanh(|x_t - \theta|/2), and its expectation is M(θ)=E[tanh⁡(|x−θ|/2)]M(\theta) = \mathrm E[\tanh(|x - \theta|/2)]. So θ^n=ln⁡f^n−ln⁡m\hat\theta_n = \ln\hat f_n - \ln m minimises MnM_n; a minimiser exists, because MnM_n is continuous, tends to 1 as |θ||\theta| grows, and is below 1 at each xtx_t. (setup; from C3)
  2. (a). Write κ(u)=sech2⁡(u/2)tanh⁡(|u|/2)≥0\kappa(u) = \operatorname{sech}^2(u/2)\tanh(|u|/2) \ge 0. Since κ(u)\kappa(u) is the derivative of tanh2⁡(u/2)\tanh^2(u/2) for u>0u > 0, ∫κ(u)du=2\int\kappa(u)\,du = 2. By unimodality g(ln⁡m+u)≤g(ln⁡m)g(\ln m + u) \le g(\ln m), so A=∫κ(u)g(ln⁡m+u)du≤2g(ln⁡m)A = \int\kappa(u)\,g(\ln m + u)\,du \le 2g(\ln m). A density cannot equal g(ln⁡m)g(\ln m) everywhere, and g(ln⁡m+u)g(\ln m + u) does not increase as |u||u| grows; so g(ln⁡m+u)<g(ln⁡m)g(\ln m + u) < g(\ln m) for every |u||u| beyond some point, where κ>0\kappa > 0. So the inequality is strict. (definition; from S1)
  3. The expected loss is lowest at 0 only. M(θ)M(\theta) is the expected sMAPE term of the forecast meθm\,e^\theta. By C5, M(θ)>M(0)M(\theta) > M(0) for every θ≠0\theta \ne 0. The function tanh⁡(|u|/2)\tanh(|u|/2) has slope at most 12\tfrac12, so MM and every MnM_n change by at most 12|θ1−θ2|\tfrac12|\theta_1 - \theta_2| between θ1\theta_1 and θ2\theta_2; in particular MM is continuous. And M(θ)→1>M(0)M(\theta) \to 1 > M(0) as |θ|→∞|\theta| \to \infty, by bounded convergence. So for every η>0\eta > 0 the lowest value of MM over |θ|≥η|\theta| \ge \eta is above M(0)M(0). (bounded convergence; from C5)
  4. MnM_n is close to MM for all θ\theta at once. For each θ\theta, Mn(θ)→M(θ)M_n(\theta) \to M(\theta) with probability 1 (law of large numbers; the terms are bounded). Take a grid of spacing γ\gamma on [−R,R][-R, R]: by the slope bound of step 3, once MnM_n is within γ\gamma of MM at the grid points it is within 2γ2\gamma of MM on all of [−R,R][-R, R]. Outside [−R,R][-R, R]: an actual with |xt|≤R/2|x_t| \le R/2 has tanh⁡(|xt−θ|/2)≥tanh⁡(R/4)\tanh(|x_t - \theta|/2) \ge \tanh(R/4) when |θ|>R|\theta| > R, so Mn(θ)M_n(\theta) is at least tanh⁡(R/4)\tanh(R/4) times the share of such actuals, which tends to P(|x|≤R/2)\mathrm P(|x| \le R/2). Since M(0)<1M(0) < 1, for RR large this is above M(0)M(0) by a fixed margin. (case analysis; from S3)
  5. Consistency. Mn(θ^n)≤Mn(0)M_n(\hat\theta_n) \le M_n(0), which tends to M(0)M(0). By step 4, with probability tending to 1, |θ^n|≤R|\hat\theta_n| \le R and M(θ^n)M(\hat\theta_n) is within any given margin of M(0)M(0); by step 3, |θ^n|<η|\hat\theta_n| < \eta. So θ^n→0\hat\theta_n \to 0 in probability, which is f^n→m\hat f_n \to m. This is the argument of van der Vaart (1998, Theorem 5.7). (algebra; from S4, S3, B15)
  6. The curvature at 0. By C12, step 4, with c=ln⁡mc = \ln m, the second derivative of MM at 0 is g(ln⁡m)−12A=12(2g(ln⁡m)−A)g(\ln m) - \tfrac12 A = \tfrac12(2g(\ln m) - A), which is positive by step 2. The first derivative at 0 is 0, because 0 is the minimum (step 3). (earlier result; from S2, S3, C12)
  7. Asymptotic normality. For every x≠0x \ne 0, θ↦tanh⁡(|x−θ|/2)\theta \mapsto \tanh(|x - \theta|/2) has a derivative at θ=0\theta = 0, −ψ(x)-\psi(x) with ψ(x)=12sech2⁡(x/2)\psi(x) = \tfrac12\operatorname{sech}^2(x/2) for x>0x > 0 and −12sech2⁡(x/2)-\tfrac12\operatorname{sech}^2(x/2) for x<0x < 0; and x=0x = 0 has probability zero. For every xx it changes by at most 12|θ1−θ2|\tfrac12|\theta_1 - \theta_2| (step 3). MM has a second-order expansion at its minimum 0 with a positive second derivative (step 6). And θ^n\hat\theta_n minimises MnM_n and tends to 0 in probability (step 5). These are the conditions of the theorem on estimators that minimise a criterion which is Lipschitz in the parameter (van der Vaart, 1998, Theorem 5.23). It gives that nθ^n\sqrt n\,\hat\theta_n is asymptotically normal with mean 0 and variance E[ψ(x)2]\mathrm E[\psi(x)^2] divided by the square of the second derivative of MM at 0. (cited theorem; from S3, S6, S5, B15)
  8. The variance. E[ψ(x)2]=14E[sech4⁡(x/2)]=B/4\mathrm E[\psi(x)^2] = \tfrac14\mathrm E[\operatorname{sech}^4(x/2)] = B/4, and the square of the second derivative is 14(2g(ln⁡m)−A)2\tfrac14(2g(\ln m) - A)^2 (step 6). So the variance is B/(2g(ln⁡m)−A)2B/(2g(\ln m) - A)^2. (algebra; from S6)
  9. The sample median. The logarithm of the sample median of the yty_t is the sample median of the ln⁡yt\ln y_t, which minimises the mean of |xt−θ||x_t - \theta|. The same theorem applies with |x−θ||x - \theta| in place of tanh⁡(|x−θ|/2)\tanh(|x - \theta|/2): the derivative at 0 is −1-1 for x>0x > 0 and +1+1 for x<0x < 0, with square 1; the criterion changes by at most |θ1−θ2||\theta_1 - \theta_2|; and the second derivative of E|x−θ|\mathrm E|x - \theta| at 0 is 2g(ln⁡m)2g(\ln m), positive because gg is largest at ln⁡m\ln m. So the variance is 1/(4g(ln⁡m)2)1/(4g(\ln m)^2). (cited theorem; from S8, B15)
  10. (b), the ratio. Dividing the variance of step 9 by that of step 8 gives ARE=(2g(ln⁡m)−A)2/(4g(ln⁡m)2B)\mathrm{ARE} = (2g(\ln m) - A)^2/(4g(\ln m)^2 B). (algebra; from S9, S8)
  11. (c), narrow distributions. Write ln⁡y=ln⁡m+σε\ln y = \ln m + \sigma\varepsilon with ε\varepsilon standard normal, or Laplace with scale 1 and σ=b\sigma = b; then g(ln⁡m)=h(0)/σg(\ln m) = h(0)/\sigma with hh the density of ε\varepsilon, and x=σεx = \sigma\varepsilon. As σ→0\sigma \to 0, A→0A \to 0 and B→1B \to 1 by bounded convergence, and A/(2g(ln⁡m))=σA/(2h(0))→0A/(2g(\ln m)) = \sigma A/(2h(0)) \to 0. So ARE=(1−A/(2g(ln⁡m)))2/B→1\mathrm{ARE} = (1 - A/(2g(\ln m)))^2/B \to 1. (bounded convergence; from S10)
  12. (c), wide lognormal distributions. Since ∫κ=2\int\kappa = 2 (step 2), 2g(ln⁡m)−A=∫κ(u)[g(ln⁡m)−g(ln⁡m+u)]du2g(\ln m) - A = \int\kappa(u)\,[g(\ln m) - g(\ln m + u)]\,du. For the normal density, g(ln⁡m)−g(ln⁡m+u)=(1−e−u2/(2σ2))/(σ2π)g(\ln m) - g(\ln m + u) = \big(1 - e^{-u^2/(2\sigma^2)}\big)/(\sigma\sqrt{2\pi}), and u22σ2−u48σ4≤1−e−u2/(2σ2)≤u22σ2\tfrac{u^2}{2\sigma^2} - \tfrac{u^4}{8\sigma^4} \le 1 - e^{-u^2/(2\sigma^2)} \le \tfrac{u^2}{2\sigma^2}. Integrating by parts, ∫u2κ(u)du=4∫0∞usech2⁡(u/2)du=16ln⁡2\int u^2\kappa(u)\,du = 4\int_0^\infty u\operatorname{sech}^2(u/2)\,du = 16\ln 2; and ∫u4κ(u)du\int u^4\kappa(u)\,du is finite. So 2g(ln⁡m)−A=8ln⁡2/(σ32π)+O(σ−5)2g(\ln m) - A = 8\ln 2/(\sigma^3\sqrt{2\pi}) + O(\sigma^{-5}). In the same way, with ∫sech4⁡(u/2)du=8/3\int\operatorname{sech}^4(u/2)\,du = 8/3, B=8/(3σ2π)+O(σ−3)B = 8/(3\sigma\sqrt{2\pi}) + O(\sigma^{-3}). With g(ln⁡m)=1/(σ2π)g(\ln m) = 1/(\sigma\sqrt{2\pi}), AREσ3\mathrm{ARE}\,\sigma^3 tends to 64(ln⁡2)22π⋅2π4⋅32π8=6(ln⁡2)22π\dfrac{64(\ln 2)^2}{2\pi}\cdot\dfrac{2\pi}{4}\cdot\dfrac{3\sqrt{2\pi}}{8} = 6(\ln 2)^2\sqrt{2\pi}. (algebra; from S2)
  13. (c), wide log-Laplace distributions. Here g(ln⁡m+u)=e−|u|/b/(2b)g(\ln m + u) = e^{-|u|/b}/(2b), so g(ln⁡m)−g(ln⁡m+u)=(1−e−|u|/b)/(2b)g(\ln m) - g(\ln m + u) = (1 - e^{-|u|/b})/(2b), which lies between (|u|/b−u2/(2b2))/(2b)\big(|u|/b - u^2/(2b^2)\big)/(2b) and |u|/(2b2)|u|/(2b^2). Integrating by parts, ∫|u|κ(u)du=2∫0∞sech2⁡(u/2)du=4\int |u|\,\kappa(u)\,du = 2\int_0^\infty\operatorname{sech}^2(u/2)\,du = 4. So 2g(ln⁡m)−A=2/b2+O(b−3)2g(\ln m) - A = 2/b^2 + O(b^{-3}), and in the same way B=4/(3b)+O(b−2)B = 4/(3b) + O(b^{-2}). With g(ln⁡m)=1/(2b)g(\ln m) = 1/(2b), AREb\mathrm{ARE}\,b tends to 4b4⋅b2⋅3b4⋅b=3\dfrac{4}{b^4}\cdot b^2\cdot\dfrac{3b}{4}\cdot b = 3. (algebra; from S12)
  14. So the sMAPE minimiser of a test set is asymptotically normal about the median, with a variance that the density at the median and two expectations give. On lognormal and log-Laplace actuals it is asymptotically as precise as the sample median for narrow distributions and asymptotically loses all of its precision relative to it as the actuals spread out, at the rates σ−3\sigma^{-3} and b−1b^{-1}. (conclusion; from S13)

E30 · formal · cited by C17

Proof of C17 (the first proof of the proof document A5). The Suilin loss behaves like sMAPE except near zero, where it becomes an absolute error.

  1. Which branch of the maximum. ℓ+δ≥0.5+δ\ell + \delta \ge 0.5 + \delta exactly when ℓ≥0.5\ell \ge 0.5. So the denominator of s~\tilde s is ℓ+δ\ell + \delta when ℓ≥0.5\ell \ge 0.5, and 0.5+δ0.5 + \delta when ℓ<0.5\ell < 0.5. (algebra; from D28)
  2. (a), the identity. When ℓ≥0.5\ell \ge 0.5, ℓ>0\ell > 0, so s=|y−f|/ℓs = |y-f|/\ell and |y−f|=sℓ|y-f| = s\,\ell. By step 1, s~=|y−f|/(ℓ+δ)=sℓ/(ℓ+δ)\tilde s = |y-f|/(\ell+\delta) = s\,\ell/(\ell+\delta). (algebra; from S1)
  3. (a), the factor. The function ℓ↦ℓ/(ℓ+δ)=1−δ/(ℓ+δ)\ell \mapsto \ell/(\ell+\delta) = 1 - \delta/(\ell+\delta) is increasing in ℓ\ell, below 1, and tends to 1 as ℓ\ell grows. At ℓ=0.5\ell = 0.5 it equals 0.5/0.6=5/60.5/0.6 = 5/6. So for ℓ≥0.5\ell \ge 0.5 it lies in [5/6,1)[5/6, 1). (monotonicity; from S2)
  4. (a), close to ss, but not a function of the ratio. By steps 2 and 3, 0≤s−s~=sδ/(ℓ+δ)≤δ/(ℓ+δ)0 \le s - \tilde s = s\,\delta/(\ell+\delta) \le \delta/(\ell+\delta), since s≤1s \le 1 (step 7). So s~\tilde s is within δ/(ℓ+δ)\delta/(\ell+\delta) of ss, which shrinks as ℓ\ell grows. But s~\tilde s depends on the scale: (y,f)=(1,2)(y,f) = (1,2) and (10,20)(10,20) have the same ratio and the same s=1/3s = 1/3, while s~=1/3.1\tilde s = 1/3.1 and 10/30.110/30.1. So s~\tilde s is not a function of y/fy/f alone. (algebra; from S2, S3)
  5. (a), counts. If y≥1y \ge 1, then ℓ≥1≥0.5\ell \ge 1 \ge 0.5. If y=0y = 0, then ℓ=|f|\ell = |f|, which is at least 0.5 unless |f|<0.5|f| < 0.5. So on counts only a zero actual with |f|<0.5|f| < 0.5 has ℓ<0.5\ell < 0.5. (algebra; from S4)
  6. (b). When ℓ<0.5\ell < 0.5, by step 1, s~=|y−f|/(0.5+δ)\tilde s = |y-f|/(0.5+\delta). For y=f=0y = f = 0 this is 0. For y=0y = 0 and |f|<0.5|f| < 0.5 it is |f|/(0.5+δ)|f|/(0.5+\delta). And s(0,f)=|f|/|f|=1s(0,f) = |f|/|f| = 1 for f≠0f \ne 0, the largest value of ss (step 7). (algebra; from S1)
  7. (c). |y−f|≤|y|+|f|=ℓ|y - f| \le |y| + |f| = \ell (the triangle inequality), so s≤1s \le 1. When ℓ≥0.5\ell \ge 0.5, s~≤ℓ/(ℓ+δ)<1\tilde s \le \ell/(\ell+\delta) < 1 (step 3). When ℓ<0.5\ell < 0.5, s~≤ℓ/(0.5+δ)<0.5/0.6<1\tilde s \le \ell/(0.5+\delta) < 0.5/0.6 < 1. And s~≥0\tilde s \ge 0 because it is a ratio of a non-negative number to a positive one. A quantity between 0 and 1 has an expectation between 0 and 1, whatever the joint distribution of actuals and forecasts, so the expected Suilin loss is between 0 and 200. (algebra; from S3)
  8. So the Suilin term is the sMAPE term shrunk by a factor between 5/6 and 1 wherever |y|+|f|≥0.5|y| + |f| \ge 0.5, and an absolute error where |y|+|f|<0.5|y| + |f| < 0.5: it differs from sMAPE in kind only near zero. (conclusion; from S7)

E37 · formal · cited by C20

Proof of C20 (the second proof of the proof document A5). The msMAPE term is the sMAPE term except for actuals below its constant, where a zero actual no longer gives every non-zero forecast the largest value.

  1. (a). When y≥νy \ge \nu, max⁡{y,ν}=y\max\{y,\nu\} = y, and y>0y > 0 because ν>0\nu > 0. So sν=|f−y|/(f+y)s_\nu = |f-y|/(f+y), which is s(y,f)s(y,f) for non-negative yy and ff. If every actual is at least ν\nu, the two measures are then the same function of the forecast on every test set and under every distribution of such actuals, so they have the same minimisers. (algebra; from D33)
  2. (b). When y<νy < \nu, max⁡{y,ν}=ν\max\{y,\nu\} = \nu, so sν=|f−y|/(f+ν)s_\nu = |f-y|/(f+\nu). For y=0y = 0 this is f/(f+ν)=1−ν/(f+ν)f/(f+\nu) = 1 - \nu/(f+\nu), which is 0 at f=0f = 0, increasing in ff, below 1, and tends to 1 as ff grows. The sMAPE term of a zero actual is |f|/|f|=1|f|/|f| = 1 for every f>0f > 0. (monotonicity; from S1)
  3. (c), the bounds. sν≥0s_\nu \ge 0 because its numerator is non-negative and its denominator is at least ν>0\nu > 0. Where y≥νy \ge \nu, sν=s≤1s_\nu = s \le 1, since |f−y|≤f+y|f-y| \le f+y. Where y<νy < \nu: if f≥yf \ge y, |f−y|=f−y≤f<f+ν|f-y| = f-y \le f < f+\nu; if f<yf < y, |f−y|=y−f≤y<ν≤f+ν|f-y| = y-f \le y < \nu \le f+\nu. So sν<1s_\nu < 1 there. (algebra; from S2)
  4. (c), no jump at zero. The denominator f+max⁡{y,ν}f+\max\{y,\nu\} is at least ν>0\nu > 0 for every f≥0f \ge 0, so sν(y,f)s_\nu(y,f) is a continuous function of ff on f≥0f \ge 0, at f=0f = 0 too. Since 0≤sν≤10 \le s_\nu \le 1 (step 3), the expected term is continuous in ff as well, by bounded convergence. The sMAPE term of a zero actual is 0 at f=0f = 0 and 1 for f>0f > 0; so when zero actuals have probability p0>0p_0 > 0, the expected sMAPE term jumps by p0p_0 at f=0f = 0. (bounded convergence; from S3)
  5. (d). A count is 0 or at least 1. With ν≤1\nu \le 1, a count of at least 1 is at least ν\nu, so sν=ss_\nu = s by step 1. A zero count is below ν\nu, where sν(0,f)=f/(f+ν)s_\nu(0,f) = f/(f+\nu) and s(0,f)=1s(0,f) = 1 for f>0f > 0 (step 2). (algebra; from S1, S2)
  6. So the msMAPE term is the sMAPE term for actuals of at least ν\nu. Below ν\nu its denominator no longer shrinks with the actual: a zero actual scores f/(f+ν)f/(f+\nu), and the expected msMAPE has no jump at the zero forecast. (conclusion; from S5)

Limitations

L1 · analytic flexibility · could bias magnitude · affects C13 · mitigation: none

Lab run RN1 deviated from plan PL1: RN1 ran at commit 3cb6c7c, not the plan's 92cd9a5. Between them only flatness.py and plot_flatness.py (plan PL3), test_smape.py and measures.py changed; in measures.py the unused function log_mae was removed (the supervisor dropped log-MAE from PL3, DC3). targets.py, theory.py, dists.py and the measures targets.py uses are unchanged. Reason: The commit of the PL3 revision came between PL1's plan and its run; it does not touch what PL1 runs.

L2 · analytic flexibility · could bias magnitude · affects C15 · mitigation: none

Lab run RN7 deviated from plan PL2: RN7 ran efficiency.py at commit fa0ce62, not the plan's 92cd9a5: the script runs its 16 settings in parallel processes (one per core) instead of one after another; theory.py also gained a function (balance_exact) that efficiency.py does not use. Each (setting, test-set size) draws from its own seed as before, and on a subset the parallel code and the plan's code gave byte-identical results.json. Reason: Three earlier attempts on one core (over an hour) were killed by container restarts before the lab recorded them (notes NT2.R2, NT5, NT6); in parallel the run finished in about 20 minutes. The analysis, the settings, the seeds and the results are unchanged.

L3 · sampling · reduces transferability · affects C14, C15, C21 · mitigation: none

The test sets are independent draws from a fixed distribution. Real forecast errors are often correlated and their distribution can change over a test period, so a real test set carries less information than the same number of outcomes here.

L4 · model specification · reduces transferability · affects C13, C14, C15, C16, C21 · mitigation: none

Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it.

L5 · model specification · reduces transferability · affects C6 · mitigation: none

C6 is proved for two groups that mirror each other about the median on the log scale. It is not proved for groups of different shape or spread.

L6 · measurement · reduces transferability · affects C20, C21 · mitigation: none

The definition of msMAPE is taken from Kolassa (2026, Definition 4). Smyl's article was not read. Smyl gives no value for the constant; ν = 0.1, 0.5 and 1 are this paper's choices.

L7 · other · reduces transferability · affects C19 · mitigation: partial

That the relative efficiency of finite test sets tends to the asymptotic value is not shown; the simulation of C15 agrees with it at n = 10000.

References

  1. Anderson, T. W., 1955. The integral of a symmetric unimodal function over a symmetric convex set and some probability inequalities. Proceedings of the American Mathematical Society 6, 170–176. https://doi.org/10.1090/S0002-9939-1955-0069229-1 Used by: Context
  2. Chen, C., Twycross, J., Garibaldi, J. M., 2017. A new accuracy measure based on bounded relative error for time series forecasting. PLOS ONE 12, e0174202. https://doi.org/10.1371/journal.pone.0174202 Used by: Context
  3. Goodwin, P., Lawton, R., 1999. On the asymmetry of the symmetric MAPE. International Journal of Forecasting 15, 405–408. https://doi.org/10.1016/S0169-2070(99)00007-2 Used by: Context
  4. Hewamalage, H., Bergmeir, C., Bandara, K., 2021. Recurrent Neural Networks for Time Series Forecasting: Current status and future directions. International Journal of Forecasting 37, 388–427. https://doi.org/10.1016/j.ijforecast.2020.06.008 Used by: Context, D28
  5. Hewamalage, H., Ackermann, K., Bergmeir, C., 2023. Forecast evaluation for data scientists: common pitfalls and best practices. Data Mining and Knowledge Discovery 37, 788–832. https://doi.org/10.1007/s10618-022-00894-5 Used by: Context
  6. Hyndman, R. J., Koehler, A. B., 2006. Another look at measures of forecast accuracy. International Journal of Forecasting 22, 679–688. https://doi.org/10.1016/j.ijforecast.2006.03.001 Used by: Context
  7. Kolassa, S., 2026. The Symmetric Mean Absolute Percentage Error: Unnecessary or Dangerous. Forecasting 8, 24. https://doi.org/10.3390/forecast8020024 Used by: Context, D33
  8. Makridakis, S., 1993. Accuracy measures: theoretical and practical concerns. International Journal of Forecasting 9, 527–529. https://doi.org/10.1016/0169-2070(93)90079-3 Used by: D2
  9. Smyl, S. (2025). Sparse-Proof sMAPE. Foresight: The International Journal of Applied Forecasting, 78, 29–30. https://ideas.repec.org/a/for/ijafaa/y2025i78p29-30.html Used by: Context
  10. Svetunkov, I., 2026a. Evaluating forecasts and classifiers: which measures, scoring rules and tests to use. EntryLib. entrylib:2026.000008@v2 Claims cited: C6, C33. Used by: Context
  11. Svetunkov, I., 2026b. Which forecast does each error measure reward? A simulation study of the functionals targeted by RMSE, MAE, MAPE and sMAPE. EntryLib. entrylib:2026.000009@v1 Claims cited: C5, C6, C11. Used by: Context
  12. Tofallis, C., 2015. A better measure of relative prediction accuracy for model selection and model estimation. Journal of the Operational Research Society 66, 1352–1362. https://doi.org/10.1057/jors.2014.103 Used by: Context
  13. Vaart, A. W. v. d., 1998. Asymptotic Statistics. Cambridge University Press. https://doi.org/10.1017/CBO9780511802256 Used by: C19

Artefacts

  • A1 · proof Proofs of the formal claims, version 3, in Markdown with LaTeX: section H3 with part (c); H5 shortened to the weighted median; new sections H12 (the ordinary median, with the checker's example), H13 (signal-to-noise ratios) and H14 (large-sample efficiency). Checked independently by AC6 (CK27 to CK31, CK34 to CK37) and corrected as it proposed; after its checks the notation was made consistent (no second meaning for k, a, G), which AC6 read and found to change nothing. Sections H1 to H14 prove C1, C2, C3, C9, C4, C5, C6, C7, C11, C8, C10, C12, C18 and C19, in that order. Sections H1 to H11 were also checked independently by lab actor AC4 (proof-check-report.md) and by the supervisor. proofs.md
  • A2 · proof The same proofs, version 3, rendered as PDF. proofs.pdf
  • A3 · figure · diagram Figure 1 of the proofs: the weight y/(y+f)^2 of an actual on the forecast that minimises expected sMAPE, against y/f on a log scale, relative to its largest value, beside MAE's constant weight; drawn by claimset-code/smape-target/plot_weights.py. smape_weights.png
  • A4 · other The independent checker's reports on the proofs (Claude Fable 5.2, lab actor AC4): derivations from the statements alone, then comparisons with the proofs, with its counterexamples and the corrections they led to. proof-check-report.md
  • A5 · proof Proofs of C17 (record hypothesis H41: the Suilin loss beside sMAPE; the measure renamed, content as before) and of C20 (record hypothesis H42: the msMAPE of Smyl beside sMAPE), in Markdown with LaTeX; checked independently by AC6 (CK32, CK33). The msMAPE term is written s_nu here, as in this claimset; the recorded statement writes p. In version 2 the file was proof-msmape.md. proof-suilin.md
  • A6 · data Every result of the lab record (17182 results of 17 runs) at full precision: the lab result's ID, its run, the quantity as the run's script named it, its conditions and its value (both as JSON). The evidence's tables give the values to four significant digits. smape-target.results.csv
  • A7 · lab_record Lab record of the study: questions, hypotheses, plans, runs, results and decisions, as an append-only, hash-chained event log (Claimset lab record specification v0.22, begun 2026-11-16; exported with claimset 0.44.6). lab.jsonl
  • A8 · data Every result of lab run RN1 (PL1: the targets of RMSE, MAE, MAPE and sMAPE on 34 distributions) at full precision, one row per result, sorted by setting: the lab result's ID, its run, the quantity, its conditions and its value (both as JSON). The evidence's tables give selected values to four significant digits. smape-target.RN1.results.csv
  • A9 · figure · data_plot inverse Gaussian shape=1000: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=1000.png). RN1-inverse_Gaussian_shape_1000.png
  • A10 · figure · data_plot inverse Gaussian shape=200: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=200.png). RN1-inverse_Gaussian_shape_200.png
  • A11 · figure · data_plot inverse Gaussian shape=100: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=100.png). RN1-inverse_Gaussian_shape_100.png
  • A12 · figure · data_plot inverse Gaussian shape=50: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=50.png). RN1-inverse_Gaussian_shape_50.png
  • A13 · figure · data_plot inverse Gaussian shape=20: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=20.png). RN1-inverse_Gaussian_shape_20.png
  • A14 · figure · data_plot inverse Gaussian shape=10: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/inverse_Gaussian_shape=10.png). RN1-inverse_Gaussian_shape_10.png
  • A15 · figure · data_plot log-skewed left sd=0.5 skew=-2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=0.5_skew=-2.png). RN1-log-skewed_left_sd_0.5_skew_-2.png
  • A16 · figure · data_plot log-skewed left sd=0.5 skew=-1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=0.5_skew=-1.png). RN1-log-skewed_left_sd_0.5_skew_-1.png
  • A17 · figure · data_plot log-skewed left sd=0.5 skew=-0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=0.5_skew=-0.5.png). RN1-log-skewed_left_sd_0.5_skew_-0.5.png
  • A18 · figure · data_plot log-skewed right sd=0.5 skew=+2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=0.5_skew=_2.png). RN1-log-skewed_right_sd_0.5_skew__2.png
  • A19 · figure · data_plot log-skewed right sd=0.5 skew=+1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=0.5_skew=_1.png). RN1-log-skewed_right_sd_0.5_skew__1.png
  • A20 · figure · data_plot log-skewed right sd=0.5 skew=+0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=0.5_skew=_0.5.png). RN1-log-skewed_right_sd_0.5_skew__0.5.p…
  • A21 · figure · data_plot lognormal sigma=0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5.png). RN1-lognormal_sigma_0.5.png
  • A22 · figure · data_plot log-skewed left sd=1 skew=-2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=1_skew=-2.png). RN1-log-skewed_left_sd_1_skew_-2.png
  • A23 · figure · data_plot log-skewed left sd=1 skew=-1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=1_skew=-1.png). RN1-log-skewed_left_sd_1_skew_-1.png
  • A24 · figure · data_plot log-skewed left sd=1 skew=-0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_left_sd=1_skew=-0.5.png). RN1-log-skewed_left_sd_1_skew_-0.5.png
  • A25 · figure · data_plot log-skewed right sd=1 skew=+2: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=1_skew=_2.png). RN1-log-skewed_right_sd_1_skew__2.png
  • A26 · figure · data_plot log-skewed right sd=1 skew=+1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=1_skew=_1.png). RN1-log-skewed_right_sd_1_skew__1.png
  • A27 · figure · data_plot log-skewed right sd=1 skew=+0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/log-skewed_right_sd=1_skew=_0.5.png). RN1-log-skewed_right_sd_1_skew__0.5.png
  • A28 · figure · data_plot lognormal sigma=1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1.png). RN1-lognormal_sigma_1.png
  • A29 · figure · data_plot bimodal equal modes: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/bimodal_equal_modes.png). RN1-bimodal_equal_modes.png
  • A30 · figure · data_plot bimodal left mode higher: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/bimodal_left_mode_higher.png). RN1-bimodal_left_mode_higher.png
  • A31 · figure · data_plot bimodal right mode higher: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/bimodal_right_mode_higher.png). RN1-bimodal_right_mode_higher.png
  • A32 · figure · data_plot normal straddling zero mean=0: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/normal_straddling_zero_mean=0.png). RN1-normal_straddling_zero_mean_0.png
  • A33 · figure · data_plot normal straddling zero mean=0.5: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/normal_straddling_zero_mean=0.5.png). RN1-normal_straddling_zero_mean_0.5.png
  • A34 · figure · data_plot normal straddling zero mean=1: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/normal_straddling_zero_mean=1.png). RN1-normal_straddling_zero_mean_1.png
  • A35 · figure · data_plot lognormal sigma=0.5 with zeros p0=0.25: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.25.png). RN1-lognormal_sigma_0.5_with_zeros_p0_0…
  • A36 · figure · data_plot lognormal sigma=0.5 with zeros p0=0.42: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.42.png). RN1-lognormal_sigma_0.5_with_zeros_p0_0…
  • A37 · figure · data_plot lognormal sigma=0.5 with zeros p0=0.47: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.47.png). RN1-lognormal_sigma_0.5_with_zeros_p0_0…
  • A38 · figure · data_plot lognormal sigma=0.5 with zeros p0=0.6: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=0.5_with_zeros_p0=0.6.png). RN1-lognormal_sigma_0.5_with_zeros_p0_0…
  • A39 · figure · data_plot lognormal sigma=1 with zeros p0=0.25: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.25.png). RN1-lognormal_sigma_1_with_zeros_p0_0.2…
  • A40 · figure · data_plot lognormal sigma=1 with zeros p0=0.37: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.37.png). RN1-lognormal_sigma_1_with_zeros_p0_0.3…
  • A41 · figure · data_plot lognormal sigma=1 with zeros p0=0.42: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.42.png). RN1-lognormal_sigma_1_with_zeros_p0_0.4…
  • A42 · figure · data_plot lognormal sigma=1 with zeros p0=0.6: for each measure (RMSE, MAE, MAPE, sMAPE), one violin per test-set size showing how the candidate the measure ranks first is spread over 1000 test sets, as its quantile level, with a boxplot of its median, quartiles and 5th and 95th percentiles and a diamond at its mean; a dashed line marks the measure's preferred candidate and dotted lines the other measures'. Drawn by lab run RN1 from its own results (lab/runs/RN1/figures/lognormal_sigma=1_with_zeros_p0=0.6.png). RN1-lognormal_sigma_1_with_zeros_p0_0.6…
  • A43 · data Every result of lab run RN2 (PL3: flatness of expected sMAPE as the spread of ln y grows) at full precision, one row per result, sorted by setting: the lab result's ID, its run, the quantity, its conditions and its value (both as JSON). The evidence's tables give selected values to four significant digits. smape-target.RN2.results.csv
  • A44 · figure · data_plot The quantile level of the sMAPE target against the spread sigma of ln y, for lognormal and log-Gumbel outcomes, beside the level of the log-mode: the target moves from the median to the mode of ln y as the spread grows. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_target.png). RN2-flatness_target.png
  • A45 · figure · data_plot Expected sMAPE against the quantile level of the forecast, one curve per spread sigma, for each family: the curve flattens towards 200 as sigma grows. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_expected_sMAPE.png). RN2-flatness_expected_sMAPE.png
  • A46 · figure · data_plot Lognormal outcomes: how much worse a forecast half a sigma or one sigma away from each measure's target scores, as a share of the score at the target, against sigma. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_sensitivity.png). RN2-flatness_sensitivity.png
  • A47 · figure · data_plot Ranking agreement: the share of test sets that prefer each measure's target to a forecast away from it, against sigma, for several test-set sizes. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_agreement.png). RN2-flatness_agreement.png
  • A48 · figure · data_plot Where the sMAPE minimiser of a test set falls, and how widely it spreads: its quantile level over test sets against sigma, for each family. Drawn by lab run RN2 from its own results (lab/runs/RN2/figures/flatness_minimiser.png). RN2-flatness_minimiser.png
  • A49 · data Every result of lab run RN7 (PL2: efficiency of the sMAPE minimiser against the sample median) at full precision, one row per result, sorted by setting: the lab result's ID, its run, the quantity, its conditions and its value (both as JSON). The evidence's tables give selected values to four significant digits. smape-target.RN7.results.csv
  • A50 · figure · data_plot lognormal outcomes: relative efficiency of the sMAPE minimiser of a test set against the sample median, against test-set size, one panel per spread, with its 99% band and the large-sample value as a dashed line; below 1 the sMAPE minimiser varies more than the sample median. Drawn by lab run RN7 from its own results (lab/runs/RN7/figures/efficiency_lognormal.png). RN7-efficiency_lognormal.png
  • A51 · figure · data_plot log-Laplace outcomes: relative efficiency of the sMAPE minimiser of a test set against the sample median, against test-set size, one panel per spread, with its 99% band and the large-sample value as a dashed line; below 1 the sMAPE minimiser varies more than the sample median. Drawn by lab run RN7 from its own results (lab/runs/RN7/figures/efficiency_log-Laplace.png). RN7-efficiency_log-Laplace.png
  • A52 · data Every result of lab run RN10 (PL11: the Suilin and sMAPE targets) at full precision, one row per result, sorted by setting: the lab result's ID, its run, the quantity, its conditions and its value (both as JSON). The evidence's tables give selected values to four significant digits. The run named the Suilin loss msMAPE, as version 2 of the paper did: in this file the measure "msMAPE" and the quantity "relative difference of the msMAPE target from the sMAPE target" are the Suilin loss's. smape-target.RN10.results.csv
  • A53 · figure · data_plot The Suilin loss beside sMAPE: near zero, the two targets as a share of the median against the share of zeros, for lognormal outcomes at medians 1 and 100, with the zero thresholds at median 1; away from zero, the relative difference of the Suilin target from the sMAPE target for each distribution; on Poisson counts, the two targets against the mean. The figure was drawn when the paper called the Suilin loss msMAPE, and labels it so. Drawn by lab run RN11 from the recorded results of RN10 (lab/runs/RN11/figures/msmape_targets.png). RN11-msmape_targets.png
  • A54 · code The study's code: the distributions, the exact quadrature of expected sMAPE, the formula checks, the three simulations (targets.py, efficiency.py, flatness.py), and the targets of the Suilin loss (msmape_targets.py) and of msMAPE on intermittent demand (msmape_intermittent.py). smape-target
  • A55 · data Every result of lab run RN15 (PL14: the targets of sMAPE, msMAPE and the Suilin loss on intermittent demand) at full precision, one row per result, sorted by setting: the lab result's ID, its run, the quantity, its conditions and its value (both as JSON). The evidence's tables give selected values to four significant digits. smape-target.RN15.results.csv
  • A56 · figure · data_plot The forecast with the lowest expected score (the target) against the share of zero actuals, on intermittent demand: one panel per family of settings (Poisson counts, negative binomial counts, lognormal demand sizes with zeros), one line per measure (sMAPE, msMAPE with nu = 0.1, 0.5 and 1, the Suilin loss). Drawn by lab run RN15 from its own results (lab/runs/RN15/figures/msmape_intermittent.png). RN15-msmape_intermittent.png

Score

Support of the paper: 99, Well supported

Scores are computed by rules from the claimset file (EntryLib rubric 0.7): they check the record, not the science, and may change as the rubric and reviews develop. The rubric. Every issue as text, for an agent preparing the next version.

The mean of its 13 principal claims; the lowest is C15 at 89. Open problems in the principal claims and what they rest on, each counted once, with the most it costs a claim:

  1. −6 Open problems Limitation L2 (could bias magnitude, not mitigated): Lab run RN7 deviated from plan PL2: RN7 ran efficiency.py at commit fa0ce62, not the plan's 92cd9a5: the script runs its 16 settings in parallel processes (one per core) instead of one after another; theory.py also gained a function (balance_exact) that efficiency.py does not use. Each (setting, test-set size) draws from its own seed as before, and on a subset the parallel code and the plan's code gave byte-identical results.json. Reason: Three earlier attempts on one core (over an hour) were killed by container restarts before the lab recorded them (notes NT2.R2, NT5, NT6); in parallel the run finished in about 20 minutes. The analysis, the settings, the seeds and the results are unchanged. In C15. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  2. −5 Context A principal claim related to no earlier work. In C6. To remove it: Relate the claim to the earlier work it extends, addresses or departs from.
  3. −3 Plan The lab record flags the test: code changed after plan. In C15. To remove it: Rerun the test as planned, or explain the flag in a limitation.
  4. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The test sets are independent draws from a fixed distribution. Real forecast errors are often correlated and their distribution can change over a test period, so a real test set carries less information than the same number of outcomes here. In C14, C15, C21. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  5. −1 Open problems Limitation L4 (reduces transferability, not mitigated): Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it. In C13, C14, C15, C16, C21. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  6. −1 Open problems Limitation L5 (reduces transferability, not mitigated): C6 is proved for two groups that mirror each other about the median on the log scale. It is not proved for groups of different shape or spread. In C6. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  7. −1 Open problems Limitation L6 (reduces transferability, not mitigated): The definition of msMAPE is taken from Kolassa (2026, Definition 4). Smyl's article was not read. Smyl gives no value for the constant; ν = 0.1, 0.5 and 1 are this paper's choices. In C20, C21. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

Claims

C1: 100, Well supported

Novelty: extends earlier work, by the authors

Nothing lost.

C2: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C3: 100, Well supported

Novelty: extends earlier work, by the authors

Nothing lost.

C4: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C5: 100, Well supported

Novelty: replicates earlier work, by the authors

Nothing lost.

C6: 94, Well supported

Novelty: new, by the authors

  1. −5 Context A principal claim related to no earlier work. To remove it: Relate the claim to the earlier work it extends, addresses or departs from.
  2. −1 Open problems Limitation L5 (reduces transferability, not mitigated): C6 is proved for two groups that mirror each other about the median on the log scale. It is not proved for groups of different shape or spread. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C7: 100, Well supported

Novelty: extends earlier work, by the authors

Nothing lost.

C8: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C9: 100, Well supported

Novelty: extends earlier work, by the authors

Nothing lost.

C10: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C11: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C12: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C13: 93, Well supported

Novelty: new, by the authors

  1. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): Lab run RN1 deviated from plan PL1: RN1 ran at commit 3cb6c7c, not the plan's 92cd9a5. Between them only flatness.py and plot_flatness.py (plan PL3), test_smape.py and measures.py changed; in measures.py the unused function log_mae was removed (the supervisor dropped log-MAE from PL3, DC3). targets.py, theory.py, dists.py and the measures targets.py uses are unchanged. Reason: The commit of the PL3 revision came between PL1's plan and its run; it does not touch what PL1 runs. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  2. −1 Open problems Limitation L4 (reduces transferability, not mitigated): Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C14: 93, Well supported

Novelty: new, by the authors

  1. −5 Context Numbers in the statement match no recorded result. To remove it: Record each number as a result or table cell of the evidence the claim cites, and list it in the support's results; or correct the number.
  2. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The test sets are independent draws from a fixed distribution. Real forecast errors are often correlated and their distribution can change over a test period, so a real test set carries less information than the same number of outcomes here. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  3. −1 Open problems Limitation L4 (reduces transferability, not mitigated): Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C15: 89, Well supported

Novelty: new, by the authors

  1. −6 Open problems Limitation L2 (could bias magnitude, not mitigated): Lab run RN7 deviated from plan PL2: RN7 ran efficiency.py at commit fa0ce62, not the plan's 92cd9a5: the script runs its 16 settings in parallel processes (one per core) instead of one after another; theory.py also gained a function (balance_exact) that efficiency.py does not use. Each (setting, test-set size) draws from its own seed as before, and on a subset the parallel code and the plan's code gave byte-identical results.json. Reason: Three earlier attempts on one core (over an hour) were killed by container restarts before the lab recorded them (notes NT2.R2, NT5, NT6); in parallel the run finished in about 20 minutes. The analysis, the settings, the seeds and the results are unchanged. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  2. −3 Plan The lab record flags the test: code changed after plan. To remove it: Rerun the test as planned, or explain the flag in a limitation.
  3. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The test sets are independent draws from a fixed distribution. Real forecast errors are often correlated and their distribution can change over a test period, so a real test set carries less information than the same number of outcomes here. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  4. −1 Open problems Limitation L4 (reduces transferability, not mitigated): Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C16: 99, Well supported

Novelty: new, by the authors

  1. −1 Open problems Limitation L4 (reduces transferability, not mitigated): Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C17: 100, Well supported

Novelty: extends earlier work, by the authors

Nothing lost.

C18: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C19: 100, Well supported

Novelty: new, by the authors

Nothing lost.

C20: 99, Well supported

Novelty: extends earlier work, by the authors

  1. −1 Open problems Limitation L6 (reduces transferability, not mitigated): The definition of msMAPE is taken from Kolassa (2026, Definition 4). Smyl's article was not read. Smyl gives no value for the constant; ν = 0.1, 0.5 and 1 are this paper's choices. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C21: 97, Well supported

Novelty: new, by the authors

  1. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The test sets are independent draws from a fixed distribution. Real forecast errors are often correlated and their distribution can change over a test period, so a real test set carries less information than the same number of outcomes here. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  2. −1 Open problems Limitation L4 (reduces transferability, not mitigated): Only the distribution families studied were computed; the formal claims (C4, C10) say what holds beyond them, the simulations only illustrate it. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  3. −1 Open problems Limitation L6 (reduces transferability, not mitigated): The definition of msMAPE is taken from Kolassa (2026, Definition 4). Smyl's article was not read. Smyl gives no value for the constant; ν = 0.1, 0.5 and 1 are this paper's choices. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

Not claimed

Planned proofs that give no claim of their own; the lab record has each with its reason.

  • Reworded, then proved again: H1 → C1, H2 → C2, H3 → C3, H23 → C3, H4 → C4, H33 → C4, H5 → C5, H18 → C6, H8 → C7, H12 → C8, H13 → C9, H19 → C10, H34 → C12, H35 → C17, H39 (not supported) → C18, H39 (not supported) → C18, H43 (inconclusive) → C18, H45 (inconclusive) → C18, H40 (not supported) → C19
  • Merged: H16 (inconclusive) → C4, H9 (inconclusive) → C3, H10 (inconclusive) → C2
  • Dropped: H11 (inconclusive)

Versions

  • Version 3 (this page) · 18 November 2026 at 13:08 UTC · updated
  • Version 2 · 17 November 2026 at 22:21 UTC · updated
  • Version 1 · 17 November 2026 at 15:01 UTC · active

Every version is kept in the public registry (github.com/entrylib/registry).