Ground floor, Building of Scienze Statistiche, room VII) from 18th March (h: 15.00-1830) to19th  March  (h: 9.00-12.30), 2013.

Professor: Bettina Grun (Johannes Kepler Universität Lin)

Abstract: Finite mixture models have been used for more than 100 years to model data because they constitute a flexible model class for modelling unobserved heterogeneity and finding groups in data. In this course the model class of finite mixture models is introduced and the identifiability, estimation and application of these models is covered.  The discussion of methods for estimation focuses on ML estimation based on the EM algorithm because the EM algorithm has the advantage that a range of different mixture models can be fitted involving only slight modifications of the basic algorithm. The software tools used are add-on packages from the R statistical software.

Best Regards

Francesca

------------------------------------------------------------