Share-Ratio Estimation of the Nested Multinomial Logit Model

Abstract

Recently the nested multinomial logit (NMNL) model has been proposed to remedy the “independence of irrelevant alternatives” (HA) property of the classical multinomial logit (MNL) model. The author brings the NMNL model under generalized least squares theory, thereby enhancing its estimation with large datasets containing only market shares. In this procedure one uses a pairwise approach by constructing share ratios among choice alternatives and subsets of alternatives in a partitioned choice set. An additional advantage of this method is that it generates a chi square test for IIA in the MNL submodel, which is a special case of the NMNL model.