BPDL.m可以通过神经网络进行预测和样本分类-BPDL.m through neural network forecasting and sample classification

本文用1961-2006年广西88个气象观测站的各月雨量资料，北半球500百帕月平均高度场资料(576个格点)，国家气候中心提供的部分环流特征量资料，通过EOF分解、合成分析、S-A practical estimate on the credibility formula is presented, where a piecewise linear function is taken as the approximation of the prior distribution and