Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/248984 
Year of Publication: 
2021
Series/Report no.: 
CESifo Working Paper No. 9439
Publisher: 
Center for Economic Studies and ifo Institute (CESifo), Munich
Abstract: 
We examine the profitability of personalized pricing policies that are derived using different specifications of demand in a typical retail setting with consumer-level panel data. We generate pricing policies from a variety of models, including Bayesian hierarchical choice models, regularized regressions, and classification trees using different sets of data inputs. To compare pricing policies, we employ an inverse probability weighted estimator of profits that explicitly takes into account non-random price variation and the panel nature of the data. We find that the performance of machine learning models is highly varied, ranging from a 21% loss to a 17% gain relative to a blanket couponing strategy, and a standard Bayesian hierarchical logit model achieves a 17.5% gain. Across all models purchase histories lead to large improvements in profits, but demographic information only has a small impact. We show that out-of-sample hit probabilities, a standard measure of model performance, are uncorrelated with our profit estimator and provide poor guidance towards model selection.
Subjects: 
targeting
personalization
heterogeneity
choice models
machine learning
JEL: 
C11
C33
C45
C52
D12
L11
L81
Document Type: 
Working Paper
Appears in Collections:

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.