| Biography | |
|---|---|
 Dr. Michail Tsagris Department of Economics, University of Crete, Greece |
|
| Title: Compositional data analysis wielding a folded type model | |
| Abstract: In this talk I will present a, recently suggested, folded type model that was developed for analysing compositional data, that are defined on the simplex sample space. The proposed model employs the α-transformation for compositional data and provides a new and flexible class of distributions for modelling such data. The novel model improves upon the previous α-normal in the sense that no probability is left outside the simplex. Despite the model’s rather seemingly complex structure, the use of the EM algorithm guarantees a computationally efficient maximum likelihood estimation. The model is validated through simulation studies and compared to the simple α-normal model. The simulation studies and rea data examples illustrate that the proposed model performs better in terms of capturing the data structure, when compared to the popular logistic normal distribution. It also yields more accurate estimates when compared to the α-normal model. Key words: α-transformation; EM-algorithm; compositional data; folding transformation | |
| Biography: Michail Tsagris received his BSc and MSc degrees in statistics from the Athens University of Economics and Business, Greece, while he obtained his PhD degree in statistics from the University of Nottingham. He is an assistant professor at the Department of Economics of UoC. Dr Tsagris has published more than 40 papers in journals, conference proceedings and book chapters. His current research interests include feature selection algorithms, computational statistics, compositional, and directional data analysis. | |
