ETS model selection principles

A technically strong model can still mislead if validation leaks future information. This brief explores ets model selection principles and the checks needed for credible comparison.

Practitioner lens

When testing ets model selection principles, compare point accuracy with decision loss. The forecast that minimizes an average error may not minimize stockout, obsolescence or capacity costs.

First experiment

Report uncertainty, compute cost and maintenance burden alongside point accuracy.

This is an original educational brief. It is a starting point for discussion, not a claim that one technique fits every supply chain.