QUADRATIC DISCRIMINATION THROUGH ORTHOGONAL TRANSFORMATION AND ITS APPLICATION TO LONG-RANGE FORECASTING OF DROUGHT AND EXCESSIVE RAINFALL

The method of quadratic discrimination through orthogonal transformation is introduced to optimize quadratic discriminant function, which has been proved more effective than the method of stepwise multiple discrimination. It is noteworthy that, in the case of a large number of predictors, it is better to make a preliminary choice by using stepwise multiple discrimination or stepwise regression so that the calculation can be made more stable and the effectiveness of discrimination can be improved.