Antonio Punzo
Department of Economics and Business
University of Catania
Advances in the use of the multivariate contaminated normal distribution
in model-based clustering
The multivariate contaminated normal (MCN) distribution represents a simple elliptical heavy-tailed generalization of the multivariate normal (MN) distribution aiming at handling and detecting mild outliers, sometimes also referred to as ‘bad’ points in the literature about the MCN. Advantageously, the two additional parameters have an interpretation of practical interest as a proportion of good observations and degree of contamination. In the talk, I will present a review of the uses of MCN distribution, and of some of its extensions, in clustering based on mixture models.
Bio: Antonio Punzo attained his Ph.D. in Methodological and Applied Statistics from the University of Milano-Bicocca, Milan, Italy. Currently, he holds the position of Full Professor of Statistics within the Department of Economics and Business at the University of Catania, Catania, Italy. Notably, he has authored numerous articles, primarily focusing on clustering and classification using mixture models, which have been published in prestigious journals and book chapters. Additionally, he serves as an Associate Editor for esteemed publications including the Journal of Classification, Statistical Methods & Applications, Biometrical Journal, and Italian Journal of Applied Statistics.
Wednesday, 4/10/24
4:00 PM
AUST 163
Webex: https://uconn-cmr.webex.com/uconn-cmr/j.php?MTID=m34325939cc6b949f01fb967dce4f6bc9
Coffee will be served at 3:30 pm in the Noether Lounge (AUST 326)
For more information, contact: Tracy Burke at tracy.burke@uconn.edu