Optimal pricing decisions under demand uncertainty: A Bayesian Mixture Model approach

Kirthi Kalyanam, Purdue University

Abstract

In using an estimated demand function to evaluate the impact of pricing decisions one has to contend with the fact that the true demand function is unknown. Often, many demand functions with different pricing prescriptions can be found to be consistent with the data, leading to confusion rather than a firm conclusion about the appropriate pricing action. In this thesis, I propose a Bayesian Mixture Model (BMM) approach to evaluate the impact of pricing decisions when there is uncertainty about demand. The approach involves: (1) The decision maker proposing a set of demand specifications, priors on the specifications and parameters, and his/her utility structure; (2) Estimating and computing posterior probabilities for the specifications using Bayesian techniques; (3) Combining the estimated specifications and posterior probabilities with the utility structure to arrive at optimal pricing decisions. Empirical implementations using two store level scanner data bases on ground coffee and bathroom tissue sales demonstrates the value of the BMM approach in formulating pricing strategy by simultaneously resolving uncertainty about competitive interactions, cannibalization and the profit maximizing price.

Degree

Ph.D.

Advisors

Hanson, Purdue University.

Subject Area

Marketing|Statistics

COinS