The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
383
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biometrics Section
|
Abstract - #304251 |
Title:
|
PFC Models with a Heterogenous Error Structure
|
Author(s):
|
Kofi Placid Adragni*+
|
Companies:
|
University of Maryland Baltimore County
|
Address:
|
302 Cedar Run Place, Catonsville, MD, 21228, United States
|
Keywords:
|
Sufficient dimension reduction ;
Inverse regression ;
Variable selection ;
Grassmann manifold ;
High dimensionality ;
Multivariate analysis
|
Abstract:
|
Principal fitted components models (PFC; Cook, RD (2007): Fisher Lecture. Dimension Reduction in Regression. Statistical Science, Vol. 22, No. 1, 1-26.) are likelihood-based methods for dimension reduction in regression. Given a p-vector predictor X and a response Y, PFC models aim at obtaining a sufficient reduction R(X) of dimension less than p, that retains all the regression information of Y contained in X. When the number of predictors p is large or larger than the number of observations n, the initial development of PFC does not allow to model predictors that, conditionally on the response, are dependent. We expand the applicability of PFC models with an heterogenous structure of the error in the model of X|Y to allow a conditional dependency among some predictors when p is large. The sufficient reduction was obtained using an efficient Grassmann manifold optimization procedure.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.