The Observation Theory EncyclopediaFrom TSKAboutBy kindBy chapterBy Lean fileLedgerProvenance

correction concept

DefinitionAn erratum or retraction, a change to a published number or claim that the record keeps beside the original, naming what was wrong and what replaced it. The Nadeau and Bengio adjustment is a statistical correction and has its own entry. Also corrected.
ExampleThe uncorrected t was stored as a sigma, and the corrected table changed nine wins to three, with seventeen ties.
BookData Mining as Observation, draft 0.2, commit f3914f0; entry id correction, kind concept.
Statusno ledger row names this entry. Corrections: none recorded.
Defining equationnone
Assumptions and scope
  • An erratum or retraction that the record keeps beside the original. A changed digest proves a changed file, so a correction is itself a sealed object, and the rows it corrects stay visible with the correction pointing at them.
  • A correction can be the wrong shape. The uncorrected t was stored as a sigma, and the fitted correction that transferred with mean absolute error zero was the identity. The rescaling of a variance for a dependence the harness ignored is a different object, the Nadeau and Bengio correction.
Prior artnone recorded
Evidencelean/DataMiningAsObservation/Seal.lean
Reviewedsemantic review 2026-09-06; generated 2026-09-10 from records at the commits on the provenance page.
predictionsealed3f2a…measurementrun9c1e…verdictreportedb7d0…each file's digest is recorded before the next step, and a changed digest proves a changed file
An erratum kept beside the original, itself sealed.

Equation

none

Conditions

Conditions are curated in entries.toml rather than read from a record.

Ledger

none

First stated

Chapter 2 section 2.4 and chapter 8 section 8.5 of Data Mining as Observation, with the standing corrections in geometric-observation/claims/LEDGER.md.

Measurements

Where the book states it Numbers, as the book’s sources table records them Source
chapter 6 section 6.4 the uncorrected t stored as sigma, 251 and 15.84, nine wins to three, seven, eight, 3 wins 17 ties 11 losses on 31 datasets, the three-way inconsistency constraint-gap/review/FINDINGS.md:1-35; constraint-gap/README.md:57-76

Failures and corrections

none

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/Seal.lean, theorems changed_of_hash_ne, hash_eq_of_eq, exists_collision, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation primer S, chapters 0, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14.

Related

Nadeau and Bengio correction; ledger class; sealed; harness.

See also

Ledger rows that cite the entry’s records without naming it: NEG-9, NEG-13 → resolved.

Sources-table rows that share a record with the entry without naming it: chapter 6 section 6.4, chapter 6 section 6.5, chapter 8 section 8.5.

Status

Generated 2026-09-10 by encyclopedia/generate.py; book at observation-data-mining f3914f0; the commit of every record is listed in the encyclopedia’s provenance.

← controlcorrelation →