ON THE OVERSPECIFICATION OF MULTINOMIAL AND NESTED LOGIT-MODELS DUE TO ALTERNATIVE SPECIFIC CONSTANTS

Discrete choice models as demand forecasting techniques have been used for transportation applications for more than thirty years. The multi nomial and nested legit models are probably the most widely applied in this context. Alternative specific constants, although playing an imp ortant role in these models, have received very little attention in th eoretical studies. In. this paper, we try to fill this gap by providin g an analysis of the overspecification caused by alternative specific constants to the log-likelihood function of multinomial and nested Leg it models. The theoretical results lead directly to a particular strat egy of alternative specific constant specification, called here the or thogonal strategy. The analysis of the relationship between arty two a rbitrary strategies and the derivation of an interesting geometrical p roperty of the orthogonal strategy provide a motivation to prefer the latter.