|
Activity Number:
|
545
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #310207 |
|
Title:
|
Variable Selection in Factor Analysis Regression and Its Application
|
|
Author(s):
|
Zenaida F. Mateo*+
|
|
Companies:
|
University of Manitoba
|
|
Address:
|
Dept of Statistics, Winnipeg, MB, R3T 2N2, Canada
|
|
Keywords:
|
backward procedure ; factor analysis regression ; variable selection ; ordinary least squares regression ; observed and predicted values
|
|
Abstract:
|
Factor Analysis (FA) regression is a statistical method of linear prediction when the dependent variables are subject to measurement errors. In this study, a backward elimination procedure is proposed for the selection of variables in FA regression. The squared correlation coefficient r squared (B-hat) was used as a criterion to measure the adequacy of a regression model and for evaluation and comparison of subset regression models. Some numerical investigation will be presented to analyze and evaluate the performance of the proposed variable selection procedure.
|
