This note argues that a Bayesian framework is almost inescapable when specifying statistical models of the LISREL type, i.e. models involving not only latent and manifest variables but also incidental parameters. Indeed, a careful specification, making every hypothesis explicit and interpretable...

Five different identification problems in mixture models are made explicit. Necessary and sufficient relationships among these problems of identification are analyzed using the concepts of weak and strong identification. This analysis is first particularized under a normality assumption and then...