交通量时间序列ARIMA预测技术研究

实时准确的交通流量预测是智能运输系统实现的前提和关键。随着预测时间间隔的进一步缩短，交通流量的不确定性越来越强。作为时域分析方法之一的ARIMA模型，以其理论基础扎实、操作步骤规范、分析结果易于解释的优点，成为时间序列分析的主流方法。文章结合SPSS软件对该预测技术进行研究，并利用某高速公路交通量调查序列进行实证分析。

Research of ARIMA Forecasting Technology for Traffic Volume Time Series

Real - time and exact prediction of short - term traffic flow volume is the precondition and key point to ITS system. And the uncertainty of traffic volume may become increasingly stronger along with the shortening of forecasting intervals. Being a model of time domain analysis method, ARIMA has the advantages of solid theoretical basis, standard operation steps, analysis results easily explained, and has become the mainstream method of the analysis of time series. The article studies the forecasting technology with the help of the sofeware of SPSS, and an example analysis of a highway traffic volume survey series is given in the end.