Recent advances in modeling complex data with latent variables
摘要：This talk introduces joint modeling approaches for analyzing complex data with latent variables. Several statistical models, including scalar-on-imaging regression model, two-part model, and additive mean residual life model, are considered to analyze imaging data, zero inflated or semi-continuous data, and time-to-event data in the presence of latent variables. The Bayesian approach and estimating equation method are used to conduct statistical inference. Several real applications to medical and financial studies are presented.