MEAN SQUARED ERROR OF YIELD PREDICTION BY SOYGRO

Yield prediction is often one of the major intended uses of a crop sim ulation model. It is therefore important to evaluate holy well a model performs as a predictor. The purpose of this study was to evaluate an d analyze how well the SOYGRO model predicts soybean yield, using as a criterion the mean squared error of prediction (MSEP). The four targe t populations for prediction were irrigated or unirrigated plots at on e location in France, for each of two varieties. The model parameters are estimated from the irrigated plots. The estimated MSEP values are on the order of 1(t ha(-1))(2) for all the target populations. For com parison, we defined an AVERAGE model. This model uses the average obse rved irrigated yield for each cultivar as the predictor of unobserved yields. AVERAGE was a better predictor than SOYGRO for the irrigated p opulations, while SOYGRO was better for the unirrigated populations, I t seems that SOYGRO has sufficient built-in biological realism to extr apolate more reasonably than the AVERAGE model from irrigated to unirr igated conditions; however, SOYGRO does not make as effective use of t he data used for parameter estimation as does AVERAGE.