{
 "bundle_of": "C23",
 "schema_version": "0.43",
 "paper": {
  "id": "entrylib:2026.000008",
  "version": 2,
  "title": "Evaluating forecasts and classifiers: which measures, scoring rules and tests to use",
  "licence": "CC-BY-4.0",
  "status": "corrected",
  "date": "2026-11-14",
  "status_notice": {
   "date": "2026-11-14",
   "reason_code": "correction",
   "statement": "Version 1 labelled the review's findings and recommendations extension, with an extends relation to the very works cited as their evidence, which said they extend those works; they state what those works establish, so version 2 labels them synthesis (claimset v0.43) and drops those relations, keeping extension only for C16, the one rule that goes beyond its sources, which extends Gneiting and Raftery (2007); and, at the guarantor's request, it rewords C3 (RMSSE as an example of the scaling), C4 (MAE and MASE as the main measure only when the mean and median forecasts coincide), C8 (absolute-error measures avoided on intermittent series), C13 (sharpness compared only among calibrated distributions and intervals) and C24 (choose a measure for what the methods predict, even if precedent used another).",
   "issued_by": "Ivan Svetunkov (ORCID 0000-0001-7826-0281)"
  }
 },
 "extensions": {
  "review": "0.1"
 },
 "claims": [
  {
   "id": "C23",
   "statement": "The accuracy of two forecasts of the same series should be compared with the [[D28:Diebold-Mariano test]].",
   "about": {
    "subject": "D28",
    "relation": "should be used to compare",
    "object": "the accuracy of two forecasts",
    "comparator": null,
    "conditions": [
     "one series"
    ]
   },
   "type": "methodological",
   "role": "principal",
   "novelty": "synthesis",
   "support": [
    {
     "kind": "argument",
     "warrant": "W23"
    }
   ],
   "provenance": "author_stated",
   "hypothesis": {
    "lab_event": "H79",
    "outcome": "supported"
   },
   "planning": "exploratory",
   "depends_on": [
    "C53"
   ],
   "scope": {
    "data_type": "two forecasts of one series"
   },
   "ext": {
    "review": {
     "kind": "recommendation"
    }
   },
   "limitations": [
    "L5"
   ],
   "assessments": [
    {
     "by": "author",
     "strength": "high",
     "rationale": "Exploratory in the lab's terms: a literature claim, stated from the guide before its papers were read, then confirmed against them (the status of its hypothesis in the lab record); no test was planned or run. It follows by the argument of its warrant from findings that its works state, quoted from full texts or read in full by the guarantor."
    }
   ],
   "verifiability": "open",
   "verifiability_flags": []
  },
  {
   "id": "C53",
   "statement": "The [[D28:Diebold-Mariano test]] of equal accuracy of two forecasts allows forecast errors that are serially correlated and not normally distributed.",
   "about": {
    "subject": "D28",
    "relation": "allows",
    "object": "serially correlated, non-normal forecast errors",
    "comparator": null,
    "conditions": []
   },
   "type": "methodological",
   "role": "supporting",
   "novelty": "synthesis",
   "support": [
    {
     "kind": "cited",
     "reference": "B16"
    }
   ],
   "provenance": "author_stated",
   "hypothesis": {
    "lab_event": "H60",
    "outcome": "supported"
   },
   "planning": "exploratory",
   "scope": {
    "data_type": "two forecasts of one series"
   },
   "ext": {
    "review": {
     "kind": "summary"
    }
   },
   "limitations": [
    "L5",
    "L6"
   ],
   "verifiability": "open",
   "verifiability_flags": []
  }
 ],
 "definitions": [
  {
   "id": "D28",
   "term": "Diebold-Mariano test",
   "kind": "method",
   "definition": "A test of equal expected loss of two competing forecasts of the same series, based on the mean loss differential and its long-run variance.",
   "source": "B16"
  }
 ],
 "warrants": [
  {
   "id": "W23",
   "reasoning": "deductive",
   "premises": [
    "C53"
   ],
   "text": "The Diebold-Mariano test allows forecast errors that are serially correlated and not normally distributed, as the errors of multi-step forecasts of one series usually are; tests that assume independent, normal errors do not hold for them."
  }
 ],
 "limitations": [
  {
   "id": "L5",
   "category": "researcher_position",
   "statement": "The claims restate EntryLib's study-planning guide on prediction, drafted by an agent of the same system that checked them here, and were stated before the works were read; the works were read to confirm what the guide states, and no search was run for work that disagrees with it.",
   "affects": [
    "C1",
    "C2",
    "C3",
    "C4",
    "C5",
    "C6",
    "C7",
    "C8",
    "C9",
    "C10",
    "C11",
    "C12",
    "C13",
    "C14",
    "C15",
    "C16",
    "C17",
    "C18",
    "C19",
    "C20",
    "C21",
    "C22",
    "C23",
    "C24",
    "C25",
    "C26",
    "C27",
    "C28",
    "C29",
    "C30",
    "C31",
    "C32",
    "C33",
    "C34",
    "C35",
    "C36",
    "C37",
    "C38",
    "C39",
    "C40",
    "C41",
    "C42",
    "C43",
    "C44",
    "C45",
    "C46",
    "C47",
    "C48",
    "C49",
    "C50",
    "C51",
    "C52",
    "C53"
   ],
   "impact": "could_bias_magnitude",
   "mitigation": "none"
  },
  {
   "id": "L6",
   "category": "data_quality",
   "statement": "These claims rest in part on works read in full by the guarantor, not by the agent that wrote the claimset (lab note NT14); no passage of those works is quoted, so a reader checks these claims against the works themselves.",
   "affects": [
    "C30",
    "C33",
    "C40",
    "C41",
    "C45",
    "C47",
    "C48",
    "C49",
    "C52",
    "C53"
   ],
   "impact": "reduces_precision",
   "mitigation": "none"
  }
 ],
 "references": [
  {
   "id": "B16",
   "target": "10.1080/07350015.1995.10524599",
   "id_type": "doi",
   "ext": {
    "review": {
     "read": "full_text",
     "checked_on": "2026-11-13",
     "lab_event": "SC18"
    }
   }
  }
 ]
}
