Choosing Customized Marketing Offers to Attract a Desirable Mix of Customers
35 Pages Posted: 28 Jul 2010
Date Written: July 1, 2010
Abstract
This paper studies the marketing problem of a firm trying to obtain an attractive mix of customers by making customized offers. The firm values the average features of the acquired set of customers, makes decisions with uncertainty about whether and which customers will accept offers, is risk averse, and faces a resource constraint. These features do not allow the application of standard procedures, so we develop methods tailored to the stochastic nature of the problem in order to be computationally feasible. These methods include calculating probabilistic guarantees for the approximate utility function and bounds on the optimal value. Our approach overcomes both the Optimizer's Curse and the computational difficulty of enumeration techniques for this problem. We apply the approach to an MBA admission process and implement adjusted scholarship decisions. We demonstrate the methodology can generate statistically significant improvements over current management decisions. By comparing our solution to what management would do on its own, we provide insight into the nature of mistakes management made in this complex decision environment. We believe that the methodology proposed here, i.e., the model, the probabilistic guarantees, and the computational approach, will prove to be valuable in a variety of different applications.
Keywords: Choice Sets, College Choice, Utility on Averages, Statistical Approximation, Non-Convex Optimization
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
By Timo Kuosmanen and Mika Kortelainen
-
Data Envelopment Analysis as Nonparametric Least Squares Regression
By Timo Kuosmanen and Andrew L Johnson
-
Nonparametric Estimation of Concave Production Technologies by Entropic Methods
By Gad Allon, Steve Hackman, ...
-
By Andrew L Johnson and Timo Kuosmanen
-
Costs of Reducing Greenhouse Gas Emissions: A Case Study of India's Power Generation Sector
By Manish Gupta
-
A More Efficient Algorithm for Convex Nonparametric Least Squares
By Chia-yen Lee, Andrew L Johnson, ...