Case Study: Demand Forecasting for Retail & Logistics
SKU×store quantile forecasts that feed inventory decisions, reconciled across a hierarchy
Model interview answer for a retail replenishment forecasting problem (50,000 SKUs by 500 stores, daily, 28 days ahead): frame the work as an inventory decision rather than a point forecast, treat stockouts as censoring of demand, climb a forecast-value-added ladder from seasonal naive and ETS to a global LightGBM with Tweedie loss, compare four ways of handling censored history, produce quantile and lead-time-sum forecasts, convert them to order-up-to levels with the newsvendor critical ratio and its cost asymmetry, reconcile a store/region/total hierarchy with bottom-up, WLS and MinT and check coherence, run a rolling-origin backtest with horizon-wise metrics and leakage checks, size a batch architecture from explicit arithmetic, and monitor bias, FVA and drift. All numbers come from a fixed-seed synthetic panel with known true demand.
Practice questions (5)
-
View →
Your Backtest Prefers the Model That Ignores Stockouts
Advanced · Free -
View →
Which Quantile Do You Order For a 95 Percent Service Level?
Advanced · Free -
View →
Forecasting a SKU With No Sales History
Advanced -
View →
Bottom-Up, WLS or MinT for the Planning Hierarchy
Advanced -
View →
How Often Do You Retrain the Forecasting Model?
Advanced