基于ARIMA模型的时间序列建模算法和实证分析

通过对时间序列ARIMA模型建模方法的研究,将方差分析运用于时间序列建模,对季节数据做方差检验并确定周期。基于统计软件SAS分析ARIMA模型建模方法的具体算法,绘制详细的建模流程图。从模型的识别、参数估计、建模和预测等各方面介绍了模型建立和预测的全过程。利用SAS软件,结合引入的方差检验方法和算法流程对1990年1月至2010年12月的中国消费者价格指数季节性时间序列建立了乘积ARIMA模型,预测并分析了CPI的基本走势。

Modeling algorithm and empirical analysis based on the time series of the ARIMA model

Through the study of time series ARIMA model modeling method,this paper applies variance analysis to time series modeling,after which carries out relevant variance tests on season datas and finally ascertains their cycle.Based on the detailed specific algorithm of statistical software SAS on analysing the ARIMA model modeling methods,as well as its specific steps on drafting particular flow chart.This paper elaborates the overall process of the model establishment and its forecast from those various aspects such as the model identification,the parameter estimation and the modeling establishment and its forecast.Finally,it uses the SAS software which combines with the incoming variance testing method and the algorithm process to establish the product ARIMA model on Chinese consumer price index of the seasonal time sequence from January 1990 to December 2010,forecast and analyze the basic trend of the CPI.