用于时间序列的预测，包含序列特征描述、平稳性检验、序列周期判断、季节因子提取、指数平滑预测、及ARIMA预测-Sequence features for time series prediction, including the descr iption of the stationary test, to determine the sequence cycle, seasonal factor extraction, exponential smoothing, and ARIMA fore

为建立完善的ARMA模型，进行的预处理，包括野点的分离去除，离散傅里叶变换去除周期项，以及正态性、平稳性检验-For the establishment of a perfect ARMA model, the pretreatment, including the separation of the wild point removal, discrete Fourier transform to remove the periodic term, as well as the normal