office: 229 Simon Hall
phone: (314) 935-4657
fax: (314) 935-6359
He teaches statistics and econometrics to students in the MBA, specialized MS, and doctoral programs.In his research, which is available here, he has developed novel and original inferential approaches and methods for diverse problems spanning the analysis of binary, categorical and censored data, the Metropolis-Hastings algorithm, the computation of the marginal likelihood in parametric and non-parametric Bayesian models, and techniques for estimating and comparing complex models, such as univariate and multivariate models of stochastic volatilty, univariate and multivariate ARMA models, hidden Markov models, models with multiple change points, discretely observed diffusions, multivariate models of count data, causal models with endogenous treatments in cross-sectional and panel domains, and hierarchical models for longitudinal data with correlated effects, that that have been widely used by researchers in statistics, economics, social sciences and the sciences.
Current research is concerned with the development of moment-based Bayesian nonparametric inferential methods; Bayesian causal inference from moments; endogeneity testing in distribution-free regression; scaleable estimation of discrete choice models with unknown choice sets; nonparametric slope factors for asset pricing; and consistent change-point detection in non-conjugate VAR models.