Using the empirical orthogonal function(EOF)and the optimum subset regression(OSR),the statistical forecast model of dryness and wetness was established. The model includs both the principal feature of predictant and the optimum grouping of predictor.

• 利用自然正交函数（EOF）和最优子集回归（OSR）两种统计方法建立了统计预报模型，既抓住了预报对象的主要特征，又考虑了预报成因的最优组合。