SHRINKAGE ESTIMATORS APPLIED TO PREDICTION OF FRENCH WINTER-WHEAT YIELD

The paper shows that it is possible to pool departmental information t o predict regional wheat yield. In an MSEP (mean squared error of pred iction) context, shrinkage estimation in linear agrometeorological mod els improves regional prediction. Cross-validation is not precise enou gh to enable comparison between models. The authors calculate the MSEP using the hypothesis of a multinormal distribution in simple cases (t he distribution parameters are estimated with data input) and simulati ons are used for complex cases. Stein-like estimators are shown to be a superior alternative to the conventional estimators usually applied for prediction.