Computationally feasible bounds for partially observed Markov decision processes
Operations Research
A survey of algorithmic methods for partially observed Markov decision processes
Annals of Operations Research
Planning and acting in partially observable stochastic domains
Artificial Intelligence
Managing Advertising and Promotion for Long-Run Profitability
Marketing Science
Planning Marketing-Mix Strategies in the Presence of Interaction Effects
Marketing Science
Bias and Variance Approximation in Value Function Estimates
Management Science
A Partially Observed Markov Decision Process for Dynamic Pricing
Management Science
Dynamic Catalog Mailing Policies
Management Science
Modeling Online Browsing and Path Analysis Using Clickstream Data
Marketing Science
A Hidden Markov Model of Customer Relationship Dynamics
Marketing Science
Marketing Science
Value-function approximations for partially observable Markov decision processes
Journal of Artificial Intelligence Research
Incremental pruning: a simple, fast, exact method for partially observable Markov decision processes
UAI'97 Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence
IEEE Transactions on Information Theory
Portfolio Dynamics for Customers of a Multiservice Provider
Management Science
A Conjoint Model of Quantity Discounts
Marketing Science
Incorporating Direct Marketing Activity into Latent Attrition Models
Marketing Science
A Joint Model of Usage and Churn in Contractual Settings
Marketing Science
Hi-index | 0.00 |
The U.S. pharmaceutical industry spent upwards of $18 billion on marketing drugs in 2005; detailing and drug sampling activities accounted for the bulk of this spending. To stay competitive, pharmaceutical managers need to maximize the return on these marketing investments by determining which physicians to target as well as when and how to target them. In this paper, we present a two-stage approach for dynamically allocating detailing and sampling activities across physicians to maximize long-run profitability. In the first stage, we estimate a hierarchical Bayesian, nonhomogeneous hidden Markov model to assess the short-and long-term effects of pharmaceutical marketing activities. The model captures physicians' heterogeneity and dynamics in prescription behavior. In the second stage, we formulate a partially observable Markov decision process that integrates over the posterior distribution of the hidden Markov model parameters to derive a dynamic marketing resource allocation policy across physicians. We apply the proposed approach in the context of a new drug introduction by a major pharmaceutical firm. We identify three prescription-behavior states, a high degree of physicians' dynamics, and substantial long-term effects for detailing and sampling. We find that detailing is most effective as an acquisition tool, whereas sampling is most effective as a retention tool. The optimization results suggest that the firm could increase its profits substantially while decreasing its marketing spending. Our suggested framework provides important implications for dynamically managing customers and maximizing long-run profitability.