Font Size: 

The talk will present model based clustering methods to classify units based on so called multivariate mixed type longitudinal or panel data. The multivariate aspect of data points to the situation when more than one outcome is observed for each unit within a~longitudinal study at each measurement occassion. The mixed type data then arise in situations when such multivariate outcomes are not necessarily of the same type, e.g., some of them are numeric, some of them categorical. The talk will provide an overview of clustering approaches for such data developed by author over past about 10 years, partly in cooperation with Lenka Komárková, Jan Vávra, Bettina Grün and Gertraud Malsiner-Walli.