inite mixture models have become very popular tools in modeling unobserved heterogeneity (vide, e.g., McLachlan and Peel (2000), Dias and Willekens (2005)). This research extends the application of these models to sequential discrete data by incorporating misclassi