Department of Statistics Seminar
North Carolina State University
 
Presents
 
Bing Li
Penn State
 
 
Dimension reduction for non-elliptically distributed 
predictors: second-order methods
 
(joint work with Yuexiao Dong)
 
ABSTRACT
 

Many classical dimension reduction methods - especially those based on inverse conditional moments - require the predictors to have elliptical distributions, or at least to satisfy a linearity condition.  Such conditions, however, are too strong for some applications.  Li and Dong (2008) introduced the notion of the central solution space and used it to modify the first-order methods, such as Sliced Inverse Regression, so that they no longer rely on these conditions.  In this talk we generalize this idea to the second-order methods, such as Sliced Average Variance Estimator and Directional Regression.  In doing so we demonstrate that the central solution space is a versatile framework: we can use it to modify essentially all inverse conditional moment based methods to relax the distributional assumption on the predictors.  Simulation studies and an application show a substantial improvement of the modified methods over their classical counterparts.

Friday, September 19, 2008

3:35 pm-4:35 pm

321 Riddick

Refreshements will be served in the Riddick Reading Room at 3:00pm.   NOTE: No food or drink is allowed in any of the classrooms in Riddick Hall.