This paper explores pedestrian demand models at regional level and the development of trip generation models for walking to work using Poisson regression and Negative Binomial regression. Non-motorized transportation including walking has so far received relatively little attention in the transportation-planning field, and as a result the best technique for estimating pedestrian demand is still being explored by transportation planners, engineers, non-government research organizations, and local and regional government. This paper presents an empirical comparison of four different regression models for the estimation of pedestrian demand at the regional level and tries to find the most appropriate model, with reference to National Household Travel Survey 2001 data for the Baltimore region. The results show that Poisson regression seems to be more appropriate for pedestrian trip generation modeling in terms of Chi-square ratio test, Pseudo R2 and Akaike Information Criterion (AIC). However, R2 based on Deviance residuals and unrestricted Log-likelihood value at convergence confirmed the empirical studies that negative binomial regression is more appropriate for the over-dispersed dependent variable than Poisson regression.
Abstract