Analysis of traffic data based on ARMA-TS-GARCH finite mixture model

Based on the fact that in order to study time-series data, their auto-regression and periodic and the variance of the sequence and long-term random fluctuations must be analyzed, ARMA -TS-GARCH model based on ARMA model and APARCH model was proposed, and the construction of the model and the parameter estimation with the Los Angeles Long Beach area traffic data and diagnostic tests were analyzed. The results of this model were compared with those of the GARCH and ARMA-GARCH model. The results show that the ARMA-TS-GARCH model fits the data better than ARMA model and APARCH model. Using ARMA-TS-GARCH model to predict the traffic data set, the results are well.