Econometrics Seminar Series. "Pattern recognition for multivariate extremes"
Published: 24 September 2024
11 October 2024. Dr Phyllis Wan, Erasmus University Rotterdam
Dr Phyllis Wan, Erasmus School of Economics, Erasmus University Rotterdam
"Pattern recognition for multivariate extremes"
Friday, 11 October 2024. 15:00-16:30
Online
Abstract
There are effective tools in unsupervised statistical learning to uncover structures in multivariate data. However, in general, these tools cannot be directly applied to the study of extreme events, as they are designed to focus on the centre of the data distribution. In this talk, we review some of the techniques for analyzing multivariate extremes and focus on the adaptation of two techniques for pattern recognition, clustering analysis and sparse graphcial models.
Bio
Phyllis Wan is an Assistant Professor at the Department of Econometrics, Erasmus University Rotterdam. Her academic interests include statistical and econometric methods, quantitative risk management, high-dimensional data and machine learning. Her recent research focuses on extreme value analysis and its applications in economics, finance, meteorology and hydrology. She holds a PhD from Columbia University.
For further information, please contact business-seminar-series@glasgow.ac.uk.
We foster a positive and productive environment for seminars through our Code of conduct.
First published: 24 September 2024