SUMMARY Vbmp is an R package for Gaussian Process classification of data over multiple classes. It features multinomial probit regression with Gaussian Process priors and estimates class posterior probabilities employing fast variational approximations to the full posterior. This software also incorporates feature weighting by means of Automatic Relevance Determination. Being equipped with only...

SUMMARY. A rich class of parametric models is proposed for discrete choice data based on the scale mixtures of multivariate normal distributions. With special connections to multinomial probit, the new models can be implemented in a Bayesian framework without much diiculty. The proposed class of models can be extended to panel data where accounting for heterogeneities is needed. This is done by...

Recently, the null category noise model has been proposed as a simple and elegant solution to the problem of incorporating unlabeled data into a Gaussian process (GP) classification model. In this paper, we show how this binary likelihood model can be generalised to the multi-class setting through the use of the multinomial probit GP classifier. We present a Gibbs sampling scheme for sampling t...

We introduce methods for estimating a Bayesian multinomial probit switching model for unordered selection and response categories via a fully Gibbsian estimation strategy. We achieve this through the use of marginal data augmentation and a modified Imai/van Dyk-type prior for the covariance matrix. Compared to related work that requires a Metropolis step, this method improves computational effi...

This paper focuses on the youth’s decisions on household formation and human capital investment in further education or work experience and the factors influencing these choices. While previous studies limited their analyses to decisions concerning the living arrangements and the labour market, the choice set is extended here to take into account other alternatives. The decisions of either rema...

Virtually all U.S. states have now created voluntary cleanup programs (VCPs), offering liability relief and other incentives for responsible parties to remediate contaminated sites. We use a multinomial probit model to analyze participation in Oregon’s two VCPs. In contrast to previous VCP research, we find that these programs attract sites with significant contamination, not just clean ones. F...

Joint Decisions on Household Membership and Human Capital Accumulation of Youths The role of expected earnings and local markets This paper analyses the effects of expected earnings and local markets conditions on the behaviour of young adults with high school diplomas. Decisions to either remain in the parental home or form a new household are modelled jointly with those of either gaining work...

Is obesity the consequence of an optimally chosen lifestyle or do people consume too much relative to their long-term preferences? The latter perspective accepts that people might face self-control problems when exposed to the immediate gratification from food. We exploit unique survey data for Switzerland in multinomial logit and ordered probit regressions to study (i) the covariates of obesit...