基于小波分析-支持向量机（WA-SVM）组合预测模型方法，将原始降雨序列进行小波分析分解到不同层次，对每层分别采用支持向量机预测，最后合成原始序列的预测值。将该模型应用于实际流域月降雨量预测，并与单独支持向量机回归方法的结果进行了对比分析。

On basis of the combined model of wavelet analysis and support vector machine (WA - SVM), this paper decomposes the original rainfall series to different layers by means of wavelet analysis, then forecasts each layer by means of SVM, and finally obtains the forecasted results of the original time series with composition. This model to estimate the monthly rainfall sequence in the watershed has been applied for the basin rainfall forecast, and the data obtained by combined model are compared with the results obtained only using support vector machine(SVM). 