|
Activity Number:
|
112
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 12, 2002 : 12:00 PM to 1:50 PM
|
|
Sponsor:
|
Business & Economics Statistics Section*
|
| Abstract - #301230 |
|
Title:
|
Decision Theoretic Approaches to Selecting the Optimal Mailing Size In Direct Marketing
|
|
Author(s):
|
Martin Levy*+ and Daling Wen and Greg Allenby
|
|
Affiliation(s):
|
University of Cincinnati and GE Capital Mortgage Insurance Corporation and Ohio State University
|
|
Address:
|
PO Box 210130, Cincinnati, Ohio, 45221-0130, USA
|
|
Keywords:
|
Bayesian Decision Theory ; Square Error Loss ; Noninformative prior ; LINEX Loss ; BLINEX Loss ; marketing
|
|
Abstract:
|
We address the problem of selecting the optimal cut-off score in direct marketing. In such a context, a marketer rank-orders a list of potential respondents using a scoring function such as logistic regression. Then, a cutoff score is chosen below which mailings are not made. Maximum profit is the main decision goal, and mailing size is the decision variable. Our aim here is to illustrate decision theoretic strategies for choosing an optimal mailing size or cutoff score using a variety of loss functions. Our target is a database of catalogue sales obtained from the Direct Marketing Educational Foundation. We hypothesize that profitability can be improved by incorporating the intrinsic boundedness and asymmetry features of the direct mail problem into the loss structure via the BLINEX loss.
|
