MODELING STRUCTURE AND CHANCE IN TRANSITIONS - MIXED LATENT PARTIAL MARKOV-CHAIN MODELS

A family of finite mixture distribution models is presented which allo ws specification of basically different developmental processes in dis tinct latent subpopulations. In particular; random fluctuation between stares for one latent subpopulation can be modeled together with stab ility or coherent developmental trajectories for others. Formally, the se models are introduced within the framework of mixed latent Markov c hains with multiple indicators per occasion. Identifiability condition s which become necessary because of the random fluctuation assumption for a part of the population are discussed. Model specification and in terpretation are illustrated by an application to empirical data on th erapeutic interventions in infancy and early childhood.