Exploratory tobit factor analysis for multivariate censored data

We propose Multivariate Tobit models with a factor structure on the covaria nce matrix. Such models are particularly useful in the exploratory analysis of multivariate censored data and the identification of latent variables f rom behavioral data. The factor structure provides a parsimonious represent ation of the censored data and reduces the dimensionality of the integratio n required in evaluating the likelihood. In addition, the factor model para meters lend themselves to substantive interpretation and graphical display. The models are estimated with simulated maximum likelihood.. Applications to the prescription of pharmaceutical products and the analysis of multi-ca tegory buying behavior are provided.