A1051
Title: Latent variable models with copulas
Authors: Radu Craiu - University of Toronto (Canada) [presenting]
Robert Zimmerman - University of Toronto (Canada)
Abstract: Latent variable models are ubiquitous in the statistical modelling of dynamic systems or when the variable of interest is not directly observable, and one must rely on surrogate measurements. A copula-based generalization is presented, in which the joint distribution of a bivariate surrogate measure depends on the latent variable. A Bayesian model is discussed, and a computational algorithm is offered to sample the posterior. The method is illustrated using numerical experiments and data analysis.
