查看文章

Since the contribution by Warner [15] who studied travel demand as a problem in discrete binary choice, there has been a growing interest in the application of probabilistic choice models to the analysis of travel. The econometric properties of multinomial logit models have received particular attention (Nerlove and Press [ll] and McFadden [7]), and major applications of these to travel demand have appeared (see Charles River Associates [l] and Domencich and McFadden [2]). Of particular interest for the purposes of this paper are two recent studies which extend multinomial logit to the choice of housing or the joint choice of housing, residential location, and mode of travel. One of these was performed by Quigley [12] who used home interview data on renters to estimate a multinomial logit model of choice among rental housing submarkets in Pittsburgh. The other study by Lerman [6] applied the model to the joint …