Build one model that works across all four shops' data. Can what it learns on one help on the others?

Task

Forecast weekly units sold for every article in each shop, 13 weeks ahead.

Data

Sales history from four shops: Set A, Set B, Set C and Set D. They share no stores or products. Get the data.

Test

We hide each shop's last 13 weeks. You forecast them from the weeks before.

Final test

We score your model on a shop's data that is not in this set. So test it on a shop it never trained on.

Score

(1 − MASE) × 100. Higher is better; 0 means you miss by one typical week-to-week move. AutoARIMA averages 32.0. Beat it.

Question

Does what you learn on one shop help on another? Show when it helps and when it hurts.

Extra credit

Show that the other sets helped. Run your model twice: own data only, then with the other sets. Show both scores. If borrowing helped, say why. If it hurt, say when. We weigh this when we pick the winner.

20 % 40 % 60 % Set A AutoARIMA Set A, AutoARIMA: score 8.7 9 % Top-down (MSTL) Set A, Top-down (MSTL): score 10.2 10 % Your forecast Set A, Your forecast: place your own forecast here ? Set B AutoARIMA Set B, AutoARIMA: score 20.4 20 % Top-down (MSTL) Set B, Top-down (MSTL): score 20.4 20 % Your forecast Set B, Your forecast: place your own forecast here ? Set C AutoARIMA Set C, AutoARIMA: score 41.8 42 % Top-down (MSTL) Set C, Top-down (MSTL): score 37.7 38 % Your forecast Set C, Your forecast: place your own forecast here ? Set D AutoARIMA Set D, AutoARIMA: score 57.2 57 % Top-down (MSTL) Set D, Top-down (MSTL): score 44.9 45 % Your forecast Set D, Your forecast: place your own forecast here ? All four, equal weight AutoARIMA All four, equal weight, AutoARIMA: score 32.0 32 % Top-down (MSTL) All four, equal weight, Top-down (MSTL): score 28.3 28 % Your forecast All four, equal weight, Your forecast: place your own forecast here ? 0 = one typical weekly move

Does borrowing another shop's shape help?

Each shop borrows Set C's yearly shape, built only from weeks before its own holdout.

Set D · holdout 2026-01-19 to 2026-04-13
13-week holdout 2,000 4,000 6,000 8,000 units / week 24 Nov 2025 22 Dec 2025 19 Jan 2026 16 Feb 2026 16 Mar 2026 13 Apr 2026 Naive (same as last week) Own-only top-down (MSTL) Borrowed Set C season Actual

Borrowed Set C season: score -15.8 (own-only -52.5, naive -28.6)

Set B · holdout 2025-07-28 to 2025-10-20
13-week holdout 17,000 18,000 19,000 20,000 21,000 units / week 2 Jun 2025 30 Jun 2025 28 Jul 2025 25 Aug 2025 22 Sep 2025 20 Oct 2025 Naive (same as last week) Own-only top-down (MSTL) Borrowed Set C season Actual

Borrowed Set C season: score -129.8 (own-only 6.9, naive -2.3)

Actual Own-only top-down (MSTL) Borrowed Set C season Naive (same as last week)

It helped one store and hurt another. When does it help?