A HETEROSCEDASTIC EXTREME-VALUE MODEL OF INTERCITY TRAVEL MODE CHOICE

Estimation of disaggregate mode choice models to estimate the ridershi p share on a proposed new (or improved) intercity travel service and t o identify the modes from which existing intercity travelers will be d iverted to the new or upgraded service constitutes a critical part of evaluating alternative travel service proposals to alleviate intercity travel congestion. This paper develops a new heteroscedastic extreme value model of intercity mode choice that overcomes the 'independence of irrelevant alternatives' (IIA) properly of the commonly used multin omial legit model. The proposed model allows a more flexible cross-ela sticity structure among alternatives than the nested legit model. It i s also simple, intuitive and much less of a computational burden than the multinomial probit model. The paper discusses the non-IIA property of the heteroscedastic extreme value model and presents an efficient and accurate Gaussian quadrature technique to estimate the heterosceda stic model using the maximum likelihood method. The multinomial legit, alternative nested legit structures, and the heteroscedastic model ar e estimated to examine the impact of improved rail service on business travel in the Toronto-Montreal corridor. The nested legit structures are either inconsistent with utility maximization principles or are no t significantly better than the multinomial legit model. The heterosce dastic extreme value model, however, is found to be superior to the mu ltinomial legit model. The heteroscedastic model predicts smaller incr eases in rail shares and smaller decreases in non-rail shares than the multinomial legit in response to rail-service improvements. It also s uggests a larger percentage decrease in air share and a smaller percen tage decrease in auto share than the multinomial legit. Thus, the mult inomial legit model is likely to provide overly optimistic projections of rail ridership and revenue, and of alleviation in inter-city trave l congestion in general, and highway traffic congestion in particular. These findings point to the limitations of the multinomial legit and nested legit models in studying intercity mode choice behavior and to the usefulness of the heteroscedastic model proposed in this paper.