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:
|
61
|
Type:
|
Topic Contributed
|
Date/Time:
|
Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract - #305104 |
Title:
|
Kernel Dimension Reduction in Data-Rich Marketing Research
|
Author(s):
|
Yuexiao Dong*+
|
Companies:
|
Temple University
|
Address:
|
3140 McLaughlin Ct, Garnet Valley, PA, 19060, United States
|
Keywords:
|
Kernel trick ;
Nonlinear dimension reduction ;
Sliced inverse regression
|
Abstract:
|
In today's data-rich environments, firms have access to massive amounts of information about actual and potential customers. The availability of such information has led to great academic interest in dimension reduction techniques that can aid in identifying significant variables that increase purchases made by existing customers and attract new customers. Historically, principal components regression, partial least squares regression, canonical correlation analysis, ridge regression and sliced inverse regression have been deployed. These methods enable researchers to build parametric or semiparametric models, and relate a focal response variable to a few factors (or indexes). The shortcoming of these approaches, however, is that they can only rely on linear combinations of the regressors. In this paper, we draw upon recent developments from the machine learning literature to introduce kernel dimension reduction techniques. By projecting the regressors to a high-dimensional feature space, we are able to identify factors that are nonlinear combinations of the input variables. Simulation study and application in a marketing context show promising results.
|
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.