Analysis And Prediction Of Short-Term Traffic Flow Based On EVIEWS

Traffic flow prediction is real-time prediction. It is generally believed that the forecast time span between time t and t +1 does not exceed 15 min (or even less than 5 min) is Short-term Traffic Flow forecast. Because transport system is the ever-changing and full of uncertainty, so the long-term traffic flow prediction of the past is not applicable to the needs of real-time control and guidance system. The Short-term Traffic Flow prediction, of which the forecast is the 5-15 min time span, plays an important role in researching dynamic traffic assignment and has become the research hot spots.EVIEWS is a tool in econometric research for collecting information, estimating, testing, designing models and using models for forecasting and solving. It has many functions such as data-processing, mapping, statistical analysis, modeling, analysis, forecasting, and simulation and so on. It has a strong operational and applicability in using EVIEWS for analysis, modeling and prediction of Short-term Traffic Flow. This article study the prediction of Short-term Traffic Flow based on its characteristics, analyse the Short-term Traffic Flow and its influencing factor and the existing prediction models, and propose to use EVIEWS to establish short-term traffic flow prediction model. Introduced the EVIEWS object foundation, data processing, Autocorrelation, Partial Autocorrelation and Unit Root Test of Short-term Traffic Flow, and estimate the forcasting equation. Then using EVIEWS describe the statistics and analysis of the Short-term Traffic Flow time series, found exponential smoothing forecasting model, the short-term traffic flow time series become a smooth sequence after 1^st difference through unit root test, adjusted the various parameters, established ARIMA model of Short-term Traffic Flow time series, and analysised than on. The results show that it is simple and intuitive, efficient and high-precision time and other characteristics, that using EVIEWS to establish prediction model for short-term traffic flow, so it can serve as an enabling tool for practical short-term traffic flow analysis and forecasting.