This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 309
Type: Contributed
Date/Time: Tuesday, August 3, 2010 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Computing
Abstract - #307222
Title: Empirical Influence Functions for Robust Partial Least Squares Regression
Author(s): Asuman Seda Turkmen*+ and Nedret Billor+
Companies: The Ohio State University and Auburn University
Address: 1179 University Drive, Newark, OH, 43055, USA 221 Parker Hall, Department of Math and Stats, Auburn, AL, 36849,
Keywords: Partial least squares ; Robustness ; Influence function ; Diagnostics ; RoPLS
Abstract:

Partial least squares (PLS) method is an important statistical tool for summarizing correlated predictor variables into a smaller set of latent variables, which have the best predictive power. However, SIMPLS which is the most commonly used algorithm in PLS is known to be sensible to outlying observations since it is based on the empirical cross-covariance matrix. In the literature several robust PLS methods have been proposed to mitigate the effect of outliers in the PLS regression and RoPLS is the most recent proposed algorithm that is highly resistant to outliers. In this study, the influence functions for the SIMPLS and RoPLS estimators are derived. Furthermore these empirical influence functions are employed to detect influential observations in regression analysis.


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 2010 program




2010 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.