Bayesian evaluation of a semi-parametric binary response model

In this paper, we develop a Bayesian analysis of a semi-parametric binary choice model. The prior speciication of the functional parameter , namely the distribution function of a latent variable, is of the Dirichlet process type and the prior speciication of the Euclidean parameter , namely the coeecients of a linear combination of exogenous variables, is left arbitrary. The model identiication is ensured by x-ing the prior expectation of the functional parameter (see Mouchart et al. (1997)). Approximations for the posterior predictive distributions are obtained from two diierent sampling methods. Several questions are studied through an exploratory numerical analysis, such as the numerical convergence of the algorithms and of the methods and the general problem of contrasting semi-parametric and purely parametric speciication. more particularly from R. Cano and M. Steel. The help from J.M. Rolin went far beyond the border of friendly, vivid and fruitful discussions; the only limit is that he should in no way be taken as responsible for any remaining error.