This article introduces a Markov chain Monte Carlo (MCMC) method for sampling the parameters of a multinomial logit model from their posterior distribution. Let yi ∈ {0, . . . ,M} denote the categorical response of subject i with covariates xi = (xi1, . . . , xip) T . Let X = (x1, . . . ,xn) T denote the design matrix, and let y = (y1, . . . , yn) T . Multinomial logit models relate yi to xi th...

3 Model Specification 5 3.1 Binary choice . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.1.1 The binary probit model . . . . . . . . . . . . . . . . . . 6 3.1.2 The binary logit model . . . . . . . . . . . . . . . . . . . 7 3.2 More than two choices . . . . . . . . . . . . . . . . . . . . . . . 8 3.2.1 The multinomial probit model . . . . . . . . . . . . . . . 8 3.2.2 The multinomi...

Understanding and accommodating heterogeneity in variance (also referred to as heteroscedasticity) and taste has become a major area of research within discrete choice analysis. Both scale and taste heterogeneity can be specified as continuous or discrete, the latter can be associated with socio economic characteristics (i.e. observed heterogeneity) or it can be derived probabilistically (i.e. ...

Gaussian process prior with an appropriate likelihood function is a flexible non-parametric model for a variety of learning tasks. One important and standard task is multi-class classification, which is the categorization of an item into one of several fixed classes. A usual likelihood function for this is the multinomial logistic likelihood function. However, exact inference with this model ha...

Abstract The study of network formation is pervasive in economics, sociology, and many other fields. In this article, we model as a ‘choice’ that made by nodes to connect nodes. We these ‘choices’ using discrete-choice models, which agents choose between two or more discrete alternatives. employ the ‘repeated-choice’ (RC) formation. argue RC overcomes important limitations multinomial logit (MN...