TESTING AND PREDICTION OF TRAFFIC FLOW DYNAMICS WITH CHAOS

This paper attempts to test for the presence of chaotic structure and to make predictions for traffic flow time series data. In order to test for chaos phenomena, Hurst exponent, Lyapunov exponent, correlation dimension, and Kolmogorov entropy are calculated from the automatic traffic count records for 20 selected stations out of Miami Freeway in the United States. Our test results reveal that strong evidence of chaotic structure, rather than random, exists in the nature of traffic flow characteristics for these traffic time series data. A confined space fuzzy neighborhoods' difference (CSFND) prediction model is then developed using the concept of "spatial and temporal nearest neighbors" of trajectories in the reconstructed phase space. This CSFND model has demonstrated high prediction accuracy in capturing the short-term chaotic traffic flow dynamics. It performs better than the FND model developed by Sakawa, et al. (1998) who only examined the "temporal nearest neighbors" of trajectories in the phase space.

[1]  Masatoshi Sakawa,et al.  A Deterministic Nonlinear Prediction Method through Fuzzy Reasoning Using Neighborhoods' Difference and Its Application to Actual Time Series Data , 1998 .

[2]  C. Sparrow.  The Fractal Geometry of Nature , 1984 .

[3]  Ramazan Gençay,et al.  A statistical framework for testing chaotic dynamics via Lyapunov exponents , 1996 .

[4]  Kanwalroop Kathy Dhanda,et al.  Chaos in oil prices? Evidence from futures markets , 2001 .

[5]  G Sugihara,et al.  Distinguishing error from chaos in ecological time series. , 1990, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[6]  Yasunari Fujimoto,et al.  Local Fuzzy Reconstruction Method for Short-term Prediction on Chaotic Timeseries , 1995 .

[7]  Michael Frame,et al.  TRAFFIC FLOW THEORY AND CHAOTIC BEHAVIOR , 1989 .

[8]  B. LeBaron,et al.  Nonlinear Dynamics, Chaos, and Instability: Statistical Theory and Economic Evidence , 1991 .

[9]  Cahit Erkal,et al.  Distinguishing between stochastic and deterministic behavior in high frequency foreign exchange rate returns: Can non-linear dynamics help forecasting?☆ , 1996 .

[10]  Farmer,et al.  Predicting chaotic time series. , 1987, Physical review letters.

[11]  Paul S. Addison,et al.  ORDER AND CHAOS IN THE DYNAMICS OF VEHICLE PLATOONS , 1996 .

[12]  Ignacio Rodriguez-Iturbe,et al.  Phase-space analysis of daily streamflow : characterization and prediction , 1998 .

[13]  Henry D. I. Abarbanel,et al.  Identification and quantification of nonstationary chaotic behavior , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[14]  Dimitrios S. Dendrinos,et al.  Traffic-flow dynamics: A search for chaos , 1994 .

[15]  William A. Barnett,et al.  Robustness of nonlinearity and chaos tests to measurement error, inference method, and sample size , 1995 .

[16]  P. Grassberger,et al.  Measuring the Strangeness of Strange Attractors , 1983 .

[17]  George Sugihara,et al.  Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series , 1990, Nature.

[18]  Xiaoyan Zhang,et al.  Chaos in a dynamic model of traffic flows in an origin-destination network. , 1998, Chaos.

[19]  P. Addison,et al.  A novel nonlinear car-following model. , 1998, Chaos.

[20]  Robert C. Hilborn,et al.  Chaos And Nonlinear Dynamics: An Introduction for Scientists and Engineers , 1994 .