Title
Vine copula-based regression models
Abstract
Vine copulas can also be used to construct flexible classes of regression models which can accommodate non linear non Gaussian dependence. This will include models for univariate and bivariate responses. For this the conditional distribution of the response given the covariates will be derived from a joint vine copula model of response and covariates.
This provides a distribution regression model allowing for simple determination of conditional quantiles. The vine copula based regression models are constructed in such a way that the conditional density can be written explicitly without integration. This allows us also to develop a forward selection strategy to avoid overfitting. These approaches will be introduced and an application involving assessing risks in flight landings will be given.
Bio
The research activities of Prof. Czado center on the field of statistics and data science. Her focus lies on modeling complex dependencies including regression effects and time/space structures using vine copula based models. These allow the construction of high dimensional multivariate distributions for data including different asymmetrical dependencies for each pair of variables. Computer-aided processes are developed/optimized for selection, estimation and adaptation to complex data structures. Applications can be found in finance and insurance as well as in engineering, earth and life sciences. A number of cooperation agreements with various international scientists and industry representatives are in place. In 2019 Prof. Czado has published a text book on analyzing dependent data with vine copulas.
After studying in Göttingen, Prof. Czado received her doctorate from Cornell University in the field of Operations Research and Industrial Engineering in 1989. She then became assistant professor and, in 1995, associate professor at York University, Toronto. In 1998, she was appointed to a professorship position in Applied Mathematical Statistics at the Technical University of Munich.
Practical
The speaker is presenting in person at the Blindern campus, Niels Henrik Abels building, 8th floor.
This seminar is collaboration between Integreat Tuesday seminar series and Tuesday seminars organised by the section Statistics and Data Science at the department of Mathematics, University of Oslo.
The Tuesdays seminars series are devoted to various topics relevant for the Integreat´s research focus. Presenters from the Integreat community and beyond have 40 minutes to present, followed by a group discussion.
Seminars are open for attendance for everybody.
For those unable to attend in person: https://uio.zoom.us/j/62535744013
Point of contacts:
- Organiser and moderator: Professor Arnoldo Frigessi, University of Oslo
- Integreat Administrative Leader Maria Dikova, University of Oslo