Dataset Factory★ EST. 2026 · BEST AT 800×600 ★
★ WELCOME TO THE DATASET FACTORY ★ PUBLIC DATA → CHECKED BUILDS → HONEST BASELINES ★ NO POP-UPS ★ NO COOKIES ★ NO TRACKERS ★ HAND-MADE HTML ★ UNDER CONSTRUCTION SINCE THE NINETIES ★
Release notes · 2026-10-22 · announcement

A week-ahead price-increase task from Malaysia's daily price panel

Suitability and evaluation

Retained for retrospective model-development research with documented timing limitations. Later evaluation rows expose earlier outcomes; predictions require restricted access.

Supported uses: Retrospective train/validation model-development experiments under the documented timing and selection limitations; Final evaluation only under a protocol that isolates holdout data from development and prevents later rows influencing earlier predictions

Not supported: Claims of verified real-time or operational forecasting validity; Treating development/validation scores as untouched final-test results; Forecasting scores from unrestricted cross-row access to this artifact

Historical feature availability
assumed
Training-label timing
unverified
Data vintage
current snapshot
Row selection
unverified
Required evaluation access
chronological

Cross-row audit: recovery demonstrated. Independent operator-commissioned audit; predictions frozen before labels were opened; holdouts sealed; two AUC implementations in exact agreement

The labeled eval split is development/validation data. Isolate final holdout data from development and prevent later rows influencing earlier predictions. Published data is not a secure hidden test service.

  • join (premise, item, T) -> (same pair, T+7 days): the next row's prices are the operands of T's label
  • The labelled holdout is presumed to carry the same self-answering structure by construction; it was never opened by the audit and must be treated as compromisable by the same join class.
  • Historical feature/label availability at the declared instant is assumed from current snapshots, not verified against release vintages.
  • The current scorer passes the complete feature frame to predict_proba; prediction-time isolation is instructed but not mechanically enforced, so any framework score on these files is gameable.
  • Observing same-day rows in one current bulk file does not verify historical weekly completeness or revision timing. Missing next-week outcomes restrict the scored population.

Corrections to historical documentation

  • Public site page, Hub card and repository README must state the restricted suitability once rendered; no frozen package bytes change.

Reviewed 2026-10-24. Applies to this artifact; original data and tags are preserved.

Malaysia's PriceCatcher programme publishes daily retail prices for hundreds of items across thousands of premises. Access is not the contribution. The contribution is the weekly panel - 397,982 train / 280,468 eval / 154,279 holdout item-premise-weeks in the state of Johor, from November 2023 to October 2026 - and the declaration that the prediction instant is the end of week W: a forecast made on Sunday night about next week's shelf price.

The label is a direction call, not a level: will the weekly modal price be higher in week W+1 than in week W. Positive rates run 14.9% / 17.5% / 20.4%, and the rise of the rate across the windows is itself a signal about the period.

The baseline is strong enough to be interesting: the harness's XGBoost pass reaches 0.7520 eval AUC, twenty-one points above the best naive reference (an item-level prior at 0.5354, with momentum and mean-reversion near 0.50). Unlike tasks where persistence wins, this one rewards learning something about the item and the week.

→ Open the dataset page for numbers, files and checksums

★ THANKS FOR VISITING · BOOKMARK THIS PAGE · TELL A FRIEND ABOUT THE DATASET FACTORY · SIGN THE GUESTBOOK BEFORE YOU LEAVE ★