Nonlinear Multivariate Time-Space Threshold Vector Error Correction Model for Short Term Traffic State Prediction

We propose Time-Space Threshold Vector Error Correction (TS-TVEC) model for short term traffic state prediction. The theory of cointegration with error correction mechanism is employed in the general design of the new model TS-TVEC. An inherent connection between mathematical form of error correction model and the microscopic traffic flow theory is revealed through the transformation of the Fundamental Diagram. A threshold regime switching framework is implemented to overcome any unknown structural changes in traffic time series. Spatial cross correlated information is incorporated with piecewise linear vector error correction model. Neural Network model is also constructed in parallel in order to test the effectiveness and robustness of the new statistical model. Our empirical study shows that TS-TVEC model is an efficient tool that is capable of modeling the complexity of stochastic traffic flow processes and feasible for online traffic state prediction.