Extremes of Regression Quantiles |
18.4.2012
Seminar of mathematical statistics
We invite you to a common seminar of the Department of Theoretical Methods IMS SAS and Department of Applied Mathematics and Statistics, Faculty of Mathematics, Physics and lnformatics, Comenius University in Bratislava where Doc. RNDr. Jan Picek, CSc. (Department of Applied Mathematics, Technical University of Liberec) will present his lecture "Extremes of Regression Quantiles". The seminar will take place on Monday, April 23, 2012, at 9:50 in lecture room XII at Faculty of Mathematics, Physics and lnformatics in Bratislava.
Abstract
The contribution deals with estimators of extreme value index based on regression quantiles in the linear regression model. The regression quantiles can be seen as a possible generalization of the quantile idea. We show the approximation of the tail quantile function of errors. Following Drees (1998) we consider a class of smooth functionals of the tail quantile function as a tool for the construction of estimators in the linear regression context. Pickands, maximum likelihood and probability weighted moments estimators are illustrated on simulated and climatological data.
|