Latent class analysis identification of syndromes in Alzheimer's disease: A Bayesian approach
Cathal Walsh · Repository of the University of Ljubljana (University of Ljubljana) · 2006
Latent variable models have been used extensively in the social sciences. In this work a latent class analysis is used to identify syndromes within Alzheimer's disease. The fitting of the model is done in a Bayesian framework, and this is examined in detail here. In particular, the label switching problem is identified, and solutions presented. Graphical summaries of the posterior distribution are included.