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

Version 1 · 17 November 2026 · Management Science and Operations Research

Complex exponential smoothing

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

Authors of the original paper: PersonIvan Svetunkov (0000-0001-7826-0281), PersonNikolaos Kourentzes (0000-0003-0211-5218), PersonJohn Keith Ord

CC-BY-4.0 · Keywords: complex exponential smoothing, exponential smoothing, state space models, forecasting

Contributions

Converted by its authors from their paper: The published article (open access, CC BY)..

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.

entrylib:2026.000012@v1 · https://entrylib.org/p/2026.000012@v1

Support: 67/100 (EntryLib rubric 0.7; points lost: Evidence −20, Plan −3, Checkable 0, Context −6, Open problems −4). Checks the record, not the science.

Summary

Exponential smoothing forecasts a series by first choosing a model: with or without a trend, with or without seasonality. This paper proposes complex exponential smoothing (CES), which uses complex variables, pairing each value with its forecast error, so that one model moves smoothly between a level and a trend as its smoothing parameter changes. It shows that CES can be stationary or non-stationary depending on that parameter, and that it corresponds to an ARMA(2,2) model whose coefficients are tied together. A seasonal version handles seasonality without seasonal indices, so model selection reduces to a choice between two models. On the series of the M1, M3, tourism and M4 forecasting competitions, CES was more accurate than exponential smoothing, ARIMA and the Theta method on most error measures, though not on all, and adding it to a combination of forecasts improved the combination.

Claims

A term is defined before the claim that first uses it, and what supports a claim stands under its statement.

Terms this claim uses

CES key term
Complex exponential smoothing: a forecasting method that combines each past observation yty_t and its forecast error ete_t into the complex variable yt+iety_t + i e_t and forecasts with an exponentially weighted sum of these values, y^t+ie^t=(α0+iα1)∑j=1t−1(1+i−(α0+iα1))j−1(yt−j+iet−j),\hat{y}_t + i\hat{e}_t = (\alpha_0 + i\alpha_1) \sum_{j=1}^{t-1} (1 + i - (\alpha_0 + i\alpha_1))^{j-1} (y_{t-j} + i e_{t-j}), or recursively y^t+ie^t=(α0+iα1)(yt−1+iet−1)+(1−α0+i−iα1)(y^t−1+ie^t−1),\hat{y}_t + i\hat{e}_t = (\alpha_0 + i\alpha_1)(y_{t-1} + i e_{t-1}) + (1 - \alpha_0 + i - i\alpha_1)(\hat{y}_{t-1} + i\hat{e}_{t-1}), where α0+iα1\alpha_0 + i\alpha_1 is the complex smoothing parameter and e^t\hat{e}_t carries the information on past forecast errors.
complex smoothing parameter
The smoothing parameter of CES, the complex number α0+iα1\alpha_0 + i\alpha_1; its real part α0\alpha_0 and imaginary part α1\alpha_1 are the two real parameters estimated, and they set how much of the past observations and of the past forecast errors is carried into the forecast.

Claim C1 proved

Depending on the value of its complex smoothing parameter, CES is a stationary or a non-stationary model.

Evidence

Everything recorded for this evidence: E3 in the record.

Earlier work and other claims

  • Extends Hyndman et al. (2008). Exponential smoothing in a state space framework (ETS) gives 30 models with no, additive or multiplicative error, trend and seasonal components, among which a model is selected by an information criterion.
  • Answers Kourentzes et al. (2014). Identifying the presence and type of trend in exponential smoothing often leads to misidentification and poorly performing long-term forecasts.
  • Answers Hyndman et al. (2002). The diverse trend alternatives of exponential smoothing in a state space framework make the selection of a model difficult.
  • Answers Kolassa (2011). Selecting the most appropriate exponential smoothing model is a challenging task.
  • Answers Barrow et al. (2020). Even when the components of exponential smoothing are identified correctly, the estimated smoothing parameters and the weights they imply can give poor forecasts.
What the record holds of this claimRecord
Claim
C1 · formal · principal · extends earlier work
Evidence
E3 in the record
Support score
SupportS 65 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Terms this claim uses

M1, M3 and tourism data
The 5,315 yearly, quarterly, monthly and other time series of the M1, M3 and tourism forecasting competitions.
nonlinear trend component
The state ctc_t of the CES state-space model, which carries information on both its own previous value ct−1c_{t-1} and the previous level lt−1l_{t-1}; it does not correspond to the trend component of exponential smoothing.
seasonal CES
CES extended with a second pair of level and nonlinear trend states taken at lag mm, the seasonal period, with its own complex smoothing parameter β0+iβ1\beta_0 + i\beta_1: yt=l0,t−1+l1,t−m+εty_t = l_{0,t-1} + l_{1,t-m} + \varepsilon_t, l0,t=l0,t−1−(1−α1)c0,t−1+(α0−α1)εt,l_{0,t} = l_{0,t-1} - (1 - \alpha_1) c_{0,t-1} + (\alpha_0 - \alpha_1) \varepsilon_t, c0,t=l0,t−1+(1−α0)c0,t−1+(α0+α1)εt,c_{0,t} = l_{0,t-1} + (1 - \alpha_0) c_{0,t-1} + (\alpha_0 + \alpha_1) \varepsilon_t, l1,t=l1,t−m−(1−β1)c1,t−m+(β0−β1)εt,l_{1,t} = l_{1,t-m} - (1 - \beta_1) c_{1,t-m} + (\beta_0 - \beta_1) \varepsilon_t, c1,t=l1,t−m+(1−β0)c1,t−m+(β0+β1)εt.c_{1,t} = l_{1,t-m} + (1 - \beta_0) c_{1,t-m} + (\beta_0 + \beta_1) \varepsilon_t.
CES with automatic seasonal selection
CES fitted by auto.ces() from the smooth package 3.1.2, which chooses between seasonal CES and non-seasonal CES by AICc.
RMSSE
Root mean squared scaled error: the root mean squared forecast error over the horizon, scaled by the mean absolute first difference of the in-sample data; reported as its mean and median over the series.
MASE
Mean absolute scaled error: the mean absolute forecast error over the horizon, scaled by the mean absolute first difference of the in-sample data; reported as its mean and median over the series.
MSIS
Mean scaled interval score: the interval score of a prediction interval averaged over the horizon and scaled by the mean absolute first difference of the in-sample data; here of the 95% prediction interval, reported as its mean and median over the series.
ETS
Exponential smoothing in a state space framework with automatic selection among its models; here es() from the smooth package with model ZXZ, selecting by AICc.
ARIMA
Automatic state space ARIMA; here auto.ssarima() from the smooth package.
Theta method
The Theta method of forecasting; here thetaf() from the forecast package 8.14.
the four compared methods
CES with automatic seasonal selection, ETS, ARIMA and the Theta method, as the benchmark runs them.

Claim C3 evidence

On the M1, M3 and tourism data, CES has the lowest mean RMSSE and the lowest median RMSSE, MASE and MSIS of four methods, CES, ETS, ARIMA and the Theta method, but not the lowest mean MASE, and the highest mean MSIS.

Evidence

E2.T1 · Mean and median RMSSE, MASE and MSIS of six methods over the 5,315 series of the M1, M3 and tourism competitions
Rowthe six compared methodsRMSSE (mean)MASE (mean)MSIS (mean)RMSSE (median)MASE (median)MSIS (median)
r1CES with automatic seasonal selection1.9592.2723.4651.171.2980.869
r2ETS1.972.2632.2581.1811.3230.925
r3ARIMA2.1342.4823.3351.2711.4190.988
r4Theta method1.9652.2522.5311.2381.3770.895
r5SCUM1.8672.1462.3331.1281.2560.844
r6SCUM(noCES)1.9112.1912.2231.1431.2790.859

Everything recorded for this evidence: E2 in the record.

Elsewhere in this paper

What the record holds of this claimRecord
Claim
C3 · descriptive · principal · no earlier result found
Evidence
E2 in the record
Scope
population: time series of the M1, M3 and tourism forecasting competitions; conditions: seasonal or non-seasonal CES selected by AICc
Limitations
L1, L3
Support score
SupportS 73 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Term this claim uses

M4 data without 10 series
The 100,000 series of the M4 competition (23,000 yearly, 24,000 quarterly, 48,000 monthly, 359 weekly, 4,227 daily and 414 hourly) except series 19357, 43160, 51552, 62738, 92370, 92435, 93159, 93196, 93412 and 93491, whose influential outliers and structural breaks made ETS break down.

Claim C4 key claimevidence

On the M4 competition data, CES has the lowest mean and median MASE and RMSSE and the lowest median MSIS of four methods, CES, ETS, ARIMA and the Theta method, but not the lowest mean MSIS.

Evidence

E8.T1 · Mean and median MASE, RMSSE and MSIS of six methods on the M4 competition data, excluding 10 series
Rowthe six compared methodsMASE (mean)RMSSE (mean)MSIS (mean)MASE (median)RMSSE (median)MSIS (median)
r1CES with automatic seasonal selection2.9212.3392.3791.8031.5290.592
r2ETS3.312.6884.7511.8311.5420.632
r3ARIMA3.0372.4362.2721.9031.6060.617
r4Theta method2.982.3942.131.8651.5740.597
r5SCUM2.7372.1981.7511.7421.4750.568
r6SCUM(noCES)2.7842.2411.7981.7691.4980.583

Everything recorded for this evidence: E8 in the record.

What the record holds of this claimRecord
Claim
C4 · descriptive · principal · no earlier result found
Evidence
E8 in the record
Scope
population: M4 competition time series; not for: 10 series on which ETS broke down
Limitations
L3, L5
Support score
SupportS 83 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C5 proved

CES has an underlying ARMA(2,2) model whose coefficients ϕ1=2−α0\phi_1 = 2 - \alpha_0, ϕ2=α0+α1−2\phi_2 = \alpha_0 + \alpha_1 - 2, θ1=2−2α0+α1\theta_1 = 2 - 2\alpha_0 + \alpha_1 and θ2=3α0+α1−2−α02−α12\theta_2 = 3\alpha_0 + \alpha_1 - 2 - \alpha_0^2 - \alpha_1^2 are nonlinear functions of the same two parameters α0\alpha_0 and α1\alpha_1.

Evidence

Everything recorded for this evidence: E5 in the record.

What the record holds of this claimRecord
Claim
C5 · formal · principal · no earlier result found
Evidence
E5 in the record
Support score
SupportS 66 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C6 proved

The parameter space of the ARMA(2,2) model underlying CES differs from that of the general ARMA(2,2) model.

Evidence

Everything recorded for this evidence: E5 in the record.

What the record holds of this claimRecord
Claim
C6 · formal · principal · no earlier result found
Evidence
E5 in the record
Support score
SupportS 66 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C7 proved

The parameter space of the ARMA(2,2) model underlying CES is not restricted to stationary models.

Evidence

Everything recorded for this evidence: E5 in the record.

What the record holds of this claimRecord
Claim
C7 · formal · principal · no earlier result found
Evidence
E5 in the record
Support score
SupportS 66 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C8 argued

seasonal CES models seasonality with lag-mm level and nonlinear trend states instead of seasonal indices.

Argument

The seasonal model takes the level and nonlinear trend states at lag m instead of carrying a set of seasonal components, so there are no seasonal indices to renormalise. Since non-seasonal CES captures both level and trend series without a choice of trend form, the only choice left is whether to include the lag-m states.

What the record holds of this claimRecord
Claim
C8 · methodological · principal · no earlier result found
Support
argument
Scope
data: seasonal univariate time series; conditions: additive error, seasonal period m
Limitations
L2
Support score
SupportS 57 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C9 argued

seasonal CES has no seasonal indices that need renormalising.

Argument

The seasonal model takes the level and nonlinear trend states at lag m instead of carrying a set of seasonal components, so there are no seasonal indices to renormalise. Since non-seasonal CES captures both level and trend series without a choice of trend form, the only choice left is whether to include the lag-m states.

What the record holds of this claimRecord
Claim
C9 · methodological · principal · no earlier result found
Support
argument
Scope
data: seasonal univariate time series; conditions: additive error, seasonal period m
Support score
SupportS 57 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C10 key claimargued

Model selection for CES reduces to choosing between the seasonal and the non-seasonal model.

Argument

The seasonal model takes the level and nonlinear trend states at lag m instead of carrying a set of seasonal components, so there are no seasonal indices to renormalise. Since non-seasonal CES captures both level and trend series without a choice of trend form, the only choice left is whether to include the lag-m states.

What the record holds of this claimRecord
Claim
C10 · methodological · principal · no earlier result found
Support
argument
Scope
data: univariate time series, seasonal or non-seasonal; conditions: selection by an information criterion
Limitations
L2
Support score
SupportS 43 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Terms this claim uses

SCUM
Simple combination of univariate models: the median of the forecasts of ETS, ARIMA, the Theta method and CES.
SCUM(noCES)
The median of the forecasts of ETS, ARIMA and the Theta method: SCUM without CES.

Claim C11 evidence

On the M1, M3 and tourism data, SCUM has lower mean and median RMSSE and MASE and lower median MSIS than SCUM(noCES), but higher mean MSIS.

Evidence

Table E2.T1: shown above.

Everything recorded for this evidence: E2 in the record.

What the record holds of this claimRecord
Claim
C11 · descriptive · principal · no earlier result found
Evidence
E2 in the record
Scope
population: time series of the M1, M3 and tourism forecasting competitions
Limitations
L3
Support score
SupportS 79 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C12 key claimevidence

On the M4 data without 10 series, SCUM has lower mean and median MASE, RMSSE and MSIS than SCUM(noCES).

Evidence

Table E8.T1: shown above.

Everything recorded for this evidence: E8 in the record.

What the record holds of this claimRecord
Claim
C12 · descriptive · principal · no earlier result found
Evidence
E8 in the record
Scope
population: M4 competition time series; not for: 10 series on which ETS broke down
Limitations
L3, L5
Support score
SupportS 83 of 100. The score checks the record, not the science.
Record
This claim on the record page
To cite it
The page of this claim

Claim C13 argued

CES encompasses level processes.

Argument

With the imaginary part of the smoothing parameter equal to one CES is the local level model ETS(A,N,N); with it above one the forecasts grow exponentially, as in multiplicative trend models; and the flexibility of its trajectories lets it follow trends that the level and trend classification of exponential smoothing captures only with separate models.

What the record holds of this claimRecord
Claim
C13 · methodological · supporting · no earlier result found
Support
argument
Support score
SupportS 59 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C14 argued

CES encompasses multiplicative trend processes.

Argument

With the imaginary part of the smoothing parameter equal to one CES is the local level model ETS(A,N,N); with it above one the forecasts grow exponentially, as in multiplicative trend models; and the flexibility of its trajectories lets it follow trends that the level and trend classification of exponential smoothing captures only with separate models.

What the record holds of this claimRecord
Claim
C14 · methodological · supporting · no earlier result found
Support
argument
Support score
SupportS 59 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C15 argued

CES approximates additive trend processes well.

Argument

With the imaginary part of the smoothing parameter equal to one CES is the local level model ETS(A,N,N); with it above one the forecasts grow exponentially, as in multiplicative trend models; and the flexibility of its trajectories lets it follow trends that the level and trend classification of exponential smoothing captures only with separate models.

What the record holds of this claimRecord
Claim
C15 · methodological · supporting · no earlier result found
Support
argument
Support score
SupportS 74 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C16 proved

CES has the single-source-of-error state-space form yt=lt−1+εty_t = l_{t-1} + \varepsilon_t, lt=lt−1−(1−α1)ct−1+(α0−α1)εtl_t = l_{t-1} - (1 - \alpha_1) c_{t-1} + (\alpha_0 - \alpha_1) \varepsilon_t, ct=lt−1+(1−α0)ct−1+(α0+α1)εtc_t = l_{t-1} + (1 - \alpha_0) c_{t-1} + (\alpha_0 + \alpha_1) \varepsilon_t, with level ltl_t and nonlinear trend component ctc_t.

Evidence

Everything recorded for this evidence: E1 in the record.

What the record holds of this claimRecord
Claim
C16 · formal · supporting · no earlier result found
Evidence
E1 in the record
Support score
SupportS 71 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C17 proved

CES is stationary, with both eigenvalues of its transition matrix inside the unit circle, when α1<5−2α0\alpha_1 < 5 - 2\alpha_0, α1<1\alpha_1 < 1 and α1>1−α0\alpha_1 > 1 - \alpha_0.

Evidence

Everything recorded for this evidence: E3 in the record.

Rests on Hyndman et al. (2008), p. 38

What the record holds of this claimRecord
Claim
C17 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C18 proved

When α1>1\alpha_1 > 1, both eigenvalues of the CES transition matrix are real and one of them is greater than one.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C18 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C19 proved

When α1>1\alpha_1 > 1, CES produces a non-stationary trajectory.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C19 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(A,N,N)
The ETS model with additive error, no trend and no seasonality, fitted without model selection.

Claim C20 proved

When α1=1\alpha_1 = 1, CES is equivalent to ETS(A,N,N).

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C20 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C23 proved

CES is stable, with all eigenvalues of its discount matrix inside the unit circle, when (α0−2.5)2+α12>1.25(\alpha_0 - 2.5)^2 + \alpha_1^2 > 1.25, (α0−0.5)2+(α1−1)2>0.25(\alpha_0 - 0.5)^2 + (\alpha_1 - 1)^2 > 0.25 and (α0−1.5)2+(α1−0.5)2<1.5(\alpha_0 - 1.5)^2 + (\alpha_1 - 0.5)^2 < 1.5.

Evidence

Everything recorded for this evidence: E4 in the record.

Rests on Hyndman et al. (2008), pp. 172-173

What the record holds of this claimRecord
Claim
C23 · formal · supporting · no earlier result found
Evidence
E4 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C25 proved

The CES point forecast w′Fh−1vtw'F^{h-1}v_t follows one of five trajectories set by the complex smoothing parameter: a flat line, exponential growth, exponential decline, harmonic oscillation converging to zero, or diverging harmonic oscillation.

Evidence

Everything recorded for this evidence: E3 in the record.

Earlier work and other claims

What the record holds of this claimRecord
Claim
C25 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C26 proved

When α1=1\alpha_1 = 1, the CES point forecast is a flat line.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C26 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C27 proved

When α1>1\alpha_1 > 1, the CES point forecast grows exponentially.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C27 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C28 proved

When (4−α02)/4<α1<1(4 - \alpha_0^2)/4 < \alpha_1 < 1, the CES point forecast declines exponentially.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C28 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C29 proved

When 1−α0<α1<(4−α02)/41 - \alpha_0 < \alpha_1 < (4 - \alpha_0^2)/4, the CES point forecast is a harmonic oscillation converging to zero.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C29 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C30 proved

When 0<α1<1−α00 < \alpha_1 < 1 - \alpha_0, the CES point forecast is a diverging harmonic oscillation.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C30 · formal · supporting · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C31 argued

seasonal CES can produce seasonality whose amplitude changes nonlinearly.

Argument

Seasonal CES consists of two CES parts, one at lag 1 and one at lag m, and each part follows one of the CES forecast trajectories, so the seasonal part can grow, decline or oscillate in its own way. The non-seasonal level can become negative while the lagged level stays positive, and the two levels evolve independently, so the seasonal part can change its amplitude and shape without any change in the level of the series.

What the record holds of this claimRecord
Claim
C31 · methodological · supporting · no earlier result found
Support
argument
Support score
SupportS 65 of 100. The score checks the record, not the science.
Record
This claim on the record page

Terms this claim uses

ADF test
Augmented Dickey-Fuller test of the null hypothesis that a time series has a unit root.
KPSS test
Kwiatkowski-Phillips-Schmidt-Shin test of the null hypothesis that a time series is stationary.

Claim C32 evidence

On M3 series 1664, which graphical analysis and ADF test and KPSS test indicate is stationary, CES estimates the stationary complex smoothing parameter α0+iα1=0.99999+0.99884i\alpha_0 + i\alpha_1 = 0.99999 + 0.99884i.

Evidence

Everything recorded for this evidence: E7 in the record.

What the record holds of this claimRecord
Claim
C32 · descriptive · supporting · no earlier result found
Evidence
E7 in the record
Support score
SupportS 90 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C33 evidence

On the trended M3 series 2721, CES estimates the non-stationary complex smoothing parameter α0+iα1=1.48098+1.00346i\alpha_0 + i\alpha_1 = 1.48098 + 1.00346i.

Evidence

Everything recorded for this evidence: E7 in the record.

What the record holds of this claimRecord
Claim
C33 · descriptive · supporting · no earlier result found
Evidence
E7 in the record
Support score
SupportS 90 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

Nemenyi test
Post hoc Nemenyi test on the ranks of an error measure across series, comparing all methods at the 5% level; here rmcb() from the greybox package.

Claim C34 evidence

On the M1, M3 and tourism data, in Nemenyi tests at the 5% level, CES has the best mean rank of four methods, CES, ETS, ARIMA and the Theta method on RMSSE and on MASE.

Evidence

Everything recorded for this evidence: E2 in the record.

What the record holds of this claimRecord
Claim
C34 · descriptive · supporting · no earlier result found
Evidence
E2 in the record
Scope
conditions: Nemenyi test on ranks of the error measures, significance level 0.05
Support score
SupportS 90 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

the six compared methods
The four compared methods and the two combinations, SCUM and SCUM(noCES).

Claim C35 evidence

On the M1, M3 and tourism data, in Nemenyi tests at the 5% level SCUM is statistically significantly better than every other method of the six compared methods, including SCUM(noCES), on RMSSE, MASE and MSIS.

Evidence

Everything recorded for this evidence: E2 in the record.

What the record holds of this claimRecord
Claim
C35 · descriptive · supporting · no earlier result found
Evidence
E2 in the record
Scope
conditions: Nemenyi test on ranks of the error measures, significance level 0.05
Support score
SupportS 90 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

non-seasonal subset
The 3,262 series of the M1, M3 and tourism data for which ETS model selection by es() from the smooth package chooses a non-seasonal model.

Claim C36 evidence

On the 3262 non-seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(A,N,N), but a higher mean MSIS.

Evidence

E9.T1 · Mean and median RMSSE, MASE and MSIS of CES and five non-seasonal ETS models on the non-seasonal subset of the M1, M3 and tourism series (the seasonal and non-seasonal subsets: non-seasonal subset)
RowCES and the non-seasonal special cases of ETSRMSSE (mean)MASE (mean)MSIS (mean)RMSSE (median)MASE (median)MSIS (median)
r1CES with automatic seasonal selection2.5342.9494.2691.6541.880.863
r2ETS(A,N,N)2.8233.2343.1041.952.2280.917
r3ETS(A,A,N)2.673.0953.7851.6941.9190.907
r4ETS(A,Ad,N)2.5022.9073.7691.661.8630.854
r5ETS(M,M,N)3.1853.6984.7681.8752.090.96
r6ETS(M,Md,N)2.6913.163.5461.7731.9750.971

Everything recorded for this evidence: E9 in the record.

Elsewhere in this paper

What the record holds of this claimRecord
Claim
C36 · descriptive · supporting · no earlier result found
Evidence
E9 in the record
Limitations
L1
Support score
SupportS 74 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(A,A,N)
The ETS model with additive error, additive trend and no seasonality, fitted without model selection.

Claim C37 evidence

On the 3262 non-seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(A,A,N), but a higher mean MSIS.

Evidence

Table E9.T1: shown above.

Everything recorded for this evidence: E9 in the record.

Elsewhere in this paper

What the record holds of this claimRecord
Claim
C37 · descriptive · supporting · no earlier result found
Evidence
E9 in the record
Support score
SupportS 74 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(A,Ad,N)
The ETS model with additive error, additive damped trend and no seasonality, fitted without model selection.

Claim C38 evidence

On the 3262 non-seasonal series of the M1, M3 and tourism data, CES has a lower median RMSSE than ETS(A,Ad,N), but higher mean RMSSE, mean and median MASE, and mean and median MSIS.

Evidence

Table E9.T1: shown above.

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C38 · descriptive · supporting · no earlier result found
Evidence
E9 in the record
Support score
SupportS 74 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(M,M,N)
The ETS model with multiplicative error, multiplicative trend and no seasonality, fitted without model selection.

Claim C39 evidence

On the 3262 non-seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE, MASE and MSIS than ETS(M,M,N).

Evidence

Table E9.T1: shown above.

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C39 · descriptive · supporting · no earlier result found
Evidence
E9 in the record
Support score
SupportS 74 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(M,Md,N)
The ETS model with multiplicative error, multiplicative damped trend and no seasonality, fitted without model selection.

Claim C40 evidence

On the 3262 non-seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(M,Md,N), but a higher mean MSIS.

Evidence

Table E9.T1: shown above.

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C40 · descriptive · supporting · no earlier result found
Evidence
E9 in the record
Support score
SupportS 74 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C41 proved

The transition matrix of the CES state-space form contains the smoothing parameters.

Evidence

Everything recorded for this evidence: E1 in the record.

What the record holds of this claimRecord
Claim
C41 · formal · secondary · no earlier result found
Evidence
E1 in the record
Support score
SupportS 71 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

Akaike information criterion
A criterion for choosing among models fitted by maximum likelihood that penalises the maximised log-likelihood by the number of estimated parameters.

Claim C42 from earlier work

The Akaike information criterion of seasonal and non-seasonal CES can be computed from a Gaussian likelihood of the same form as that of additive exponential smoothing models.

Rests on Hyndman et al. (2008), p. 68

What the record holds of this claimRecord
Claim
C42 · methodological · secondary · no earlier result found
Support
cited
Support score
SupportS 66 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C43 from earlier work

The conditional variance of CES hh steps ahead, given the level and the nonlinear trend component, can be calculated from its state-space form in the same way as for pure additive ETS models.

Rests on Hyndman et al. (2008), p. 96

What the record holds of this claimRecord
Claim
C43 · formal · secondary · no earlier result found
Support
cited
Support score
SupportS 69 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C44 proved

When α1\alpha_1 is close to 1, CES behaves like simple exponential smoothing with smoothing parameter α0−1\alpha_0 - 1.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C44 · formal · secondary · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C45 proved

When α1\alpha_1 is close to 1, the real part α0\alpha_0 of the complex smoothing parameter of CES is close to 1 for a stationary series, for which the smoothing parameter of simple exponential smoothing would be close to 0.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C45 · formal · secondary · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C46 proved

When α1\alpha_1 is close to 1, the real part α0\alpha_0 of the complex smoothing parameter of CES is close to 2 for a random walk.

Evidence

Everything recorded for this evidence: E3 in the record.

What the record holds of this claimRecord
Claim
C46 · formal · secondary · no earlier result found
Evidence
E3 in the record
Support score
SupportS 75 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C47 proved

seasonal CES has an underlying SARIMA(2,0,2m+2)(2,0,0)m_m model.

Evidence

Everything recorded for this evidence: E6 in the record.

What the record holds of this claimRecord
Claim
C47 · formal · secondary · no earlier result found
Evidence
E6 in the record
Support score
SupportS 57 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C48 argued

seasonal CES is a stationary or a non-stationary model depending on the values of its complex smoothing parameters.

Argument

Non-seasonal CES has an underlying ARMA(2,2) model that is stationary or not depending on its complex smoothing parameter; by analogy, the authors state the same of the SARIMA model underlying seasonal CES. No stationarity condition is derived for seasonal CES.

What the record holds of this claimRecord
Claim
C48 · formal · secondary · no earlier result found
Support
argument
Limitations
L2
Support score
SupportS 57 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C49 argued

seasonal CES can produce all the trajectories of non-seasonal CES.

Argument

Seasonal CES consists of two CES parts, one at lag 1 and one at lag m, and each part follows one of the CES forecast trajectories, so the seasonal part can grow, decline or oscillate in its own way. The non-seasonal level can become negative while the lagged level stays positive, and the two levels evolve independently, so the seasonal part can change its amplitude and shape without any change in the level of the series.

What the record holds of this claimRecord
Claim
C49 · methodological · secondary · no earlier result found
Support
argument
Support score
SupportS 65 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C50 argued

seasonal CES can model multiplicative or other shapes of seasonality when the level of the series does not change.

Argument

Seasonal CES consists of two CES parts, one at lag 1 and one at lag m, and each part follows one of the CES forecast trajectories, so the seasonal part can grow, decline or oscillate in its own way. The non-seasonal level can become negative while the lagged level stays positive, and the two levels evolve independently, so the seasonal part can change its amplitude and shape without any change in the level of the series.

What the record holds of this claimRecord
Claim
C50 · methodological · secondary · no earlier result found
Support
argument
Support score
SupportS 65 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C51 argued

Non-seasonal CES has 5 estimated parameters: two smoothing parameters, two initial states and the error variance.

Argument

The count follows from the model equations: each complex smoothing parameter has two real parts, each pair of level and nonlinear trend states needs initial values (m of each for the lag-m pair), and the error variance is estimated.

What the record holds of this claimRecord
Claim
C51 · methodological · secondary · no earlier result found
Support
argument
Support score
SupportS 60 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C52 argued

seasonal CES has 4+2m+2+14 + 2m + 2 + 1 estimated parameters: four smoothing parameters, 2m2m initial lagged values, two initial non-seasonal states and the error variance.

Argument

The count follows from the model equations: each complex smoothing parameter has two real parts, each pair of level and nonlinear trend states needs initial values (m of each for the lag-m pair), and the error variance is estimated.

What the record holds of this claimRecord
Claim
C52 · methodological · secondary · no earlier result found
Support
argument
Support score
SupportS 60 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

the three benchmark methods
ETS, ARIMA and the Theta method, as the benchmark runs them.

Claim C53 evidence

On the M1, M3 and tourism data, in Nemenyi tests at the 5% level, not all of the differences between CES and ETS, ARIMA and the Theta method on RMSSE and MASE are statistically significant.

Evidence

Everything recorded for this evidence: E2 in the record.

Earlier work and other claims

What the record holds of this claimRecord
Claim
C53 · descriptive · secondary · no earlier result found
Evidence
E2 in the record
Scope
conditions: Nemenyi test on ranks of the error measures, significance level 0.05
Support score
SupportS 90 of 100. The score checks the record, not the science.
Record
This claim on the record page

Terms this claim uses

seasonal subset
The 2,053 series of the M1, M3 and tourism data for which ETS model selection by es() from the smooth package chooses a seasonal model.
ETS(A,N,A)
The ETS model with additive error, no trend and additive seasonality, fitted without model selection.

Claim C54 evidence

On the 2053 seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(A,N,A), but a higher mean MSIS.

Evidence

E9.T2 · Mean and median RMSSE, MASE and MSIS of CES and five seasonal ETS models on the seasonal subset of the M1, M3 and tourism series (the seasonal and non-seasonal subsets: seasonal subset)
RowCES and the seasonal special cases of ETSRMSSE (mean)MASE (mean)MSIS (mean)RMSSE (median)MASE (median)MSIS (median)
r1CES with automatic seasonal selection1.0441.1962.1880.7560.8440.877
r2ETS(A,N,A)1.1421.2972.0370.8030.8940.92
r3ETS(A,A,A)1.1351.2952.2470.8020.8770.934
r4ETS(A,Ad,A)1.0941.2452.1550.7970.8760.946
r5ETS(M,M,M)1.1711.3522.1530.8060.8751.057
r6ETS(M,Md,M)1.1031.261.9340.7890.8751.006

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C54 · descriptive · secondary · no earlier result found
Evidence
E9 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(A,Ad,A)
The ETS model with additive error, additive damped trend and additive seasonality, fitted without model selection.

Claim C55 evidence

On the 2053 seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(A,Ad,A), but a higher mean MSIS.

Evidence

Table E9.T2: shown above.

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C55 · descriptive · secondary · no earlier result found
Evidence
E9 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(M,M,M)
The ETS model with multiplicative error, multiplicative trend and multiplicative seasonality, fitted without model selection.

Claim C56 evidence

On the 2053 seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(M,M,M), but a higher mean MSIS.

Evidence

Table E9.T2: shown above.

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C56 · descriptive · secondary · no earlier result found
Evidence
E9 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Term this claim uses

ETS(M,Md,M)
The ETS model with multiplicative error, multiplicative damped trend and multiplicative seasonality, fitted without model selection.

Claim C57 evidence

On the 2053 seasonal series of the M1, M3 and tourism data, CES has lower mean and median RMSSE and MASE and lower median MSIS than ETS(M,Md,M), but a higher mean MSIS.

Evidence

Table E9.T2: shown above.

Everything recorded for this evidence: E9 in the record.

What the record holds of this claimRecord
Claim
C57 · descriptive · secondary · no earlier result found
Evidence
E9 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C58 evidence

Information criteria can effectively identify when to switch between seasonal and non-seasonal CES.

Evidence

Table E9.T2: shown above.

Everything recorded for this evidence: E9 in the record.

Argument

On the seasonal subset of the M1, M3 and tourism data, CES with its seasonal or non-seasonal form chosen by AICc has lower RMSSE and MASE than every seasonal ETS model, which the authors take to show that the criterion chooses the seasonal form when it is needed. This is an argument from the accuracy results only: no experiment compares the selection by information criteria with other ways of choosing between seasonal and non-seasonal CES.

What the record holds of this claimRecord
Claim
C58 · interpretive · secondary · no earlier result found
Evidence
E9 in the record
Support score
SupportS 80 of 100. The score checks the record, not the science.
Record
This claim on the record page

Claim C59 evidence

CES is a robust forecasting approach.

Evidence

Table E2.T1: shown above.

Everything recorded for this evidence: E2 in the record.

Evidence

Table E8.T1: shown above.

Everything recorded for this evidence: E8 in the record.

Argument

CES has lower RMSSE than ETS, ARIMA and the Theta method on the M1, M3, tourism and M4 data and lowers the errors of the combination it is included in, which the authors take as evidence that it is robust, while acknowledging that it does not suit every forecasting scenario. Robustness itself is not defined or tested: no experiment examines how the accuracy of CES varies across conditions or departures from its assumptions.

What the record holds of this claimRecord
Claim
C59 · interpretive · secondary · no earlier result found
Evidence
E2, E8 in the record
Scope
not for: forecasting scenarios for which CES is not appropriate, which the paper does not name
Support score
SupportS 90 of 100. The score checks the record, not the science.
Record
This claim on the record page

9 more definitions that no claim names (designs, data, cases) are on the record page.

Assumptions

C60 · statistical · unchecked

The error term εt\varepsilon_t of CES is additive white noise. Rested on by C16, C42, C43, D1, E1.

C61 · theoretical · untestable

There is no misspecification error, so the forecast error ete_t of CES equals the model error εt\varepsilon_t. Rested on by C5, C16, E1, E3, E4, E5, E6.

C62 · statistical · unchecked

The error term εt\varepsilon_t of CES has no autocorrelation. Rested on by C43, D1.

C63 · statistical · unchecked

The error term εt\varepsilon_t of CES has constant variance. Rested on by C43, D1.

C64 · statistical · unchecked

The error term εt\varepsilon_t of CES is normally distributed. Rested on by C42, D1.

C65 · methodological · unchecked

ETS model selection with es() from the smooth package identifies correctly which series of the M1, M3 and tourism data are seasonal. Rested on by D33, D34.

Limitations

L1 · model specification · could bias magnitude · affects C3, C36, C37, C38, C39, C40 · mitigation: none

CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series.

L2 · model specification · reduces precision · affects C8, C10, C47 · mitigation: none

The parameter restrictions under which seasonal CES is stable are not derived.

L3 · sample size · reduces transferability · affects C3, C4, C11, C12 · mitigation: none

The effect of the length of a time series on the accuracy of CES is not analysed.

L4 · model specification · reduces transferability · affects C16 · mitigation: none

CES is formulated only with an additive error term; a version with a multiplicative error term is not developed.

L5 · sampling · could bias magnitude · affects C4, C12 · mitigation: none

The 10 M4 series on which ETS broke down were excluded from the M4 comparison, so the series were selected on the failure of one of the compared methods.

Open questions

O1 · from C11

Whether CES improves the SCUM combination because it captures long-term relations better than ETS, ARIMA and the Theta method, and nonlinear trends without separating level and trend, is not established.

References

  1. Akaike, H., 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control 19, 716–723. https://doi.org/10.1109/TAC.1974.1100705 Used by: D39
  2. Assimakopoulos, V., Nikolopoulos, K., 2000. The theta model: a decomposition approach to forecasting. International Journal of Forecasting 16, 521–530. https://doi.org/10.1016/S0169-2070(00)00066-2 Used by: D11
  3. Athanasopoulos, G., Hyndman, R. J., Song, H., Wu, D. C., 2011. The tourism forecasting competition. International Journal of Forecasting 27, 822–844. https://doi.org/10.1016/j.ijforecast.2010.04.009 Used by: A3
  4. Barrow, D., Kourentzes, N., Sandberg, R., Niklewski, J., 2020. Automatic robust estimation for exponential smoothing: Perspectives from statistics and machine learning. Expert Systems with Applications 160, 113637. https://doi.org/10.1016/j.eswa.2020.113637 Used by: Context
  5. Demšar, J. (2006). Statistical comparisons of classifiers over multiple data sets. The Journal of Machine Learning Research, 7, 1–30. Used by: D38
  6. Dickey, D. A., Fuller, W. A., 1979. Distribution of the Estimators for Autoregressive Time Series with a Unit Root. Journal of the American Statistical Association 74, 427–431. https://doi.org/10.1080/01621459.1979.10482531 Used by: D40
  7. Gneiting, T., Raftery, A. E., 2007. Strictly Proper Scoring Rules, Prediction, and Estimation. Journal of the American Statistical Association 102, 359–378. https://doi.org/10.1198/016214506000001437 Used by: D16
  8. Hyndman, R. J., Khandakar, Y., 2008. Automatic Time Series Forecasting: The forecast Package for R. Journal of Statistical Software 27. https://doi.org/10.18637/jss.v027.i03 Used by: A5
  9. 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: D15
  10. Hyndman, R. J., Koehler, A. B., Snyder, R. D., Grose, S., 2002. A state space framework for automatic forecasting using exponential smoothing methods. International Journal of Forecasting 18, 439–454. https://doi.org/10.1016/S0169-2070(01)00110-8 Used by: Context, D9
  11. Hyndman, R., Koehler, A., Ord, K., Snyder, R., 2008. Forecasting with Exponential Smoothing. Springer Series in Statistics. https://doi.org/10.1007/978-3-540-71918-2 Used by: Context, C17, C21, C22, C23, C42, C43, D18, D19, D20, D21, D22, D23, D24, D27, D28, D29, D30, D31
  12. Kolassa, S., 2011. Combining exponential smoothing forecasts using Akaike weights. International Journal of Forecasting 27, 238–251. https://doi.org/10.1016/j.ijforecast.2010.04.006 Used by: Context
  13. Kourentzes, N., Petropoulos, F., Trapero, J. R., 2014. Improving forecasting by estimating time series structural components across multiple frequencies. International Journal of Forecasting 30, 291–302. https://doi.org/10.1016/j.ijforecast.2013.09.006 Used by: Context
  14. Kwiatkowski, D., Phillips, P. C. B., Schmidt, P., Shin, Y., 1992. Testing the null hypothesis of stationarity against the alternative of a unit root. Journal of Econometrics 54, 159–178. https://doi.org/10.1016/0304-4076(92)90104-Y Used by: D41
  15. Makridakis, S., Hibon, M., 2000. The M3-Competition: results, conclusions and implications. International Journal of Forecasting 16, 451–476. https://doi.org/10.1016/S0169-2070(00)00057-1 Used by: A2
  16. Makridakis, S., Andersen, A., Carbone, R., Fildes, R., Hibon, M., Lewandowski, R., Newton, J., Parzen, E., Winkler, R., 1982. The accuracy of extrapolation (time series) methods: Results of a forecasting competition. Journal of Forecasting 1, 111–153. https://doi.org/10.1002/for.3980010202 Used by: A1
  17. Makridakis, S., Spiliotis, E., Assimakopoulos, V., 2020. The M4 Competition: 100,000 time series and 61 forecasting methods. International Journal of Forecasting 36, 54–74. https://doi.org/10.1016/j.ijforecast.2019.04.014 Used by: D32, A7
  18. Makridakis, S., Spiliotis, E., Assimakopoulos, V., 2022. M5 accuracy competition: Results, findings, and conclusions. International Journal of Forecasting 38, 1346–1364. https://doi.org/10.1016/j.ijforecast.2021.11.013 Used by: D14
  19. Ord, J. K., Koehler, A. B., Snyder, R. D., 1997. Estimation and Prediction for a Class of Dynamic Nonlinear Statistical Models. Journal of the American Statistical Association 92, 1621–1629. https://doi.org/10.1080/01621459.1997.10473684 Used by: C21, C22
  20. Petropoulos, F., Svetunkov, I., 2020. A simple combination of univariate models. International Journal of Forecasting 36, 110–115. https://doi.org/10.1016/j.ijforecast.2019.01.006 Used by: D5
  21. R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/ Used by: A8
  22. Svetunkov, I., Boylan, J. E., 2020. State-space ARIMA for supply-chain forecasting. International Journal of Production Research 58, 818–827. https://doi.org/10.1080/00207543.2019.1600764 Used by: D10
  23. Svetunkov, I., Kourentzes, N., Ord, J. K., 2022. Complex exponential smoothing. Naval Research Logistics (NRL) 69, 1108–1123. https://doi.org/10.1002/nav.22074 Used by: A9
  24. https://github.com/config-i1/greybox Used by: A6
  25. https://github.com/config-i1/smooth Used by: A4

Score

Support of the paper: 67, Partly 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 12 principal claims; the lowest is C10 at 43. Open problems in the principal claims and what they rest on, each counted once, with the most it costs a claim:

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. In C1, C2, C3, C5, C6, C7, C8, C9, C10, C11, C13, C14, C15, C16, C17, C18, C19, C20, C21, C22, C23, C24, C25, C26, C27, C28, C29, C30, C31, C34, C35, C36, C37, C38, C39, C40, C41, C42, C43, C44, C45, C46, C47, C48, C49, C50, C51, C52, C53, C54, C55, C56, C57, C58, C59. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. In C1, C10, C13, C14, C15, C21, C22, C31, C32, C33, C36, C37, C38, C39, C40, C42, C49, C50, C51, C52, C54, C55, C56, C57, C58. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. In C3, C10, C13, C14, C15, C36, C37, C38, C39, C40. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  4. −5 Plan A principal claim reported without a plan recorded beforehand. In C3, C4, C8, C9, C10, C11, C12, C47, C48. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  5. −5 Context A principal claim related to no earlier work. In C2, C3, C4, C5, C6, C7, C8, C9, C10, C11, C12, C47, C48. To remove it: Relate the claim to the earlier work it extends, addresses or departs from.
  6. −6 Open problems Limitation L5 (could bias magnitude, not mitigated): The 10 M4 series on which ETS broke down were excluded from the M4 comparison, so the series were selected on the failure of one of the compared methods. In C4, C12. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  7. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. In C5, C6, C7, C16, C41, C42, C43. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  8. −3 Open problems Limitation L2 (reduces precision, not mitigated): The parameter restrictions under which seasonal CES is stable are not derived. In C8, C9, C10, C47, C48. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  9. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The effect of the length of a time series on the accuracy of CES is not analysed. In C3, C4, C11, C12. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  10. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. In C5, C6, C7, C16, C41, C43. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

Claims

C1: 65, Partly supported

Novelty: extends earlier work, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C2: 70, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −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.

C3: 73, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  3. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  4. −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.
  5. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The effect of the length of a time series on the accuracy of CES is not analysed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C4: 83, Well supported

Novelty: new, by the authors

  1. −6 Open problems Limitation L5 (could bias magnitude, not mitigated): The 10 M4 series on which ETS broke down were excluded from the M4 comparison, so the series were selected on the failure of one of the compared methods. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The effect of the length of a time series on the accuracy of CES is not analysed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C5: 66, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −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.
  3. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  4. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C6: 66, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −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.
  3. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  4. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C7: 66, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −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.
  3. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  4. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C8: 57, Weakly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −3 Open problems Limitation L2 (reduces precision, not mitigated): The parameter restrictions under which seasonal CES is stable are not derived. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C9: 57, Weakly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −3 Open problems Limitation L2 (reduces precision, not mitigated): The parameter restrictions under which seasonal CES is stable are not derived. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C10: 43, Weakly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  4. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  5. −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.
  6. −1 Open problems Limitation L2 (reduces precision, not mitigated): The parameter restrictions under which seasonal CES is stable are not derived. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C11: 79, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The effect of the length of a time series on the accuracy of CES is not analysed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C12: 83, Well supported

Novelty: new, by the authors

  1. −6 Open problems Limitation L5 (could bias magnitude, not mitigated): The 10 M4 series on which ETS broke down were excluded from the M4 comparison, so the series were selected on the failure of one of the compared methods. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −1 Open problems Limitation L3 (reduces transferability, not mitigated): The effect of the length of a time series on the accuracy of CES is not analysed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C13: 59, Weakly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C14: 59, Weakly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C15: 74, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C16: 71, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  3. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C17: 80, Well supported

Novelty: new, by the authors

  1. −20 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C18: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C19: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C20: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C21: 70, Partly supported

Novelty: new, by the authors

  1. −20 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C22: 70, Partly supported

Novelty: new, by the authors

  1. −20 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C23: 80, Well supported

Novelty: new, by the authors

  1. −20 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C24: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C25: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C26: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C27: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C28: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C29: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C30: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C31: 65, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C32: 90, Well supported

Novelty: new, by the authors

  1. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C33: 90, Well supported

Novelty: new, by the authors

  1. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C34: 90, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C35: 90, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C36: 74, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C37: 74, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C38: 74, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C39: 74, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C40: 74, Partly supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −6 Open problems Limitation L1 (could bias magnitude, not mitigated): CES produces very large errors on a few series: on M1 series 47 it fails to identify the trend and its MSIS is 523.034 against 25.590 for ETS(M,N,N), which inflates its mean MSIS, and on the non-seasonal subset it fails on two series. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C41: 71, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  3. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C42: 66, Partly supported

Novelty: new, by the authors

  1. −20 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.
  3. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  4. −1 Open problems Assumption C64 is not checked: The error term εt\varepsilon_t of CES is normally distributed. To remove it: Test the assumption, or analyse how sensitive the claim is to it.

C43: 69, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −3 Open problems Assumption C60 is not checked: The error term εt\varepsilon_t of CES is additive white noise. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  3. −1 Open problems Limitation L4 (reduces transferability, not mitigated): CES is formulated only with an additive error term; a version with a multiplicative error term is not developed. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.
  4. −1 Open problems Assumption C62 is not checked: The error term εt\varepsilon_t of CES has no autocorrelation. To remove it: Test the assumption, or analyse how sensitive the claim is to it.
  5. −1 Open problems Assumption C63 is not checked: The error term εt\varepsilon_t of CES has constant variance. To remove it: Test the assumption, or analyse how sensitive the claim is to it.

C44: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C45: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C46: 75, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C47: 57, Weakly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −3 Open problems Limitation L2 (reduces precision, not mitigated): The parameter restrictions under which seasonal CES is stable are not derived. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C48: 57, Weakly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −5 Plan A principal claim reported without a plan recorded beforehand. To remove it: Plan the measurement in the lab record or a registration before running it, or publish the lab record that shows every run.
  3. −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.
  4. −3 Open problems Limitation L2 (reduces precision, not mitigated): The parameter restrictions under which seasonal CES is stable are not derived. To remove it: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

C49: 65, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C50: 65, Partly supported

Novelty: new, by the authors

  1. −25 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C51: 60, Partly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C52: 60, Partly supported

Novelty: new, by the authors

  1. −30 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C53: 90, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

C54: 80, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C55: 80, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C56: 80, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C57: 80, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C58: 80, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).
  2. −10 Context No scope: where the claim holds is not stated. To remove it: Give the claim a scope: population, setting, period, data, conditions, exclusions.

C59: 90, Well supported

Novelty: new, by the authors

  1. −10 Evidence Rests on argument only; an unchecked proof; evidence named without the observations the claim reports; an argument from other claims; cited work only. To remove it: Support the claim with evidence the claimset records, citing the result entries or table rows it reports, or have its proof checked; for cited work, read the work in full and quote the passage the claim rests on from a captured copy (claimset extension review).

Versions

  • Version 1 (this page) · 17 November 2026 at 14:31 UTC · active

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