Dynamic Two Stage Modeling for Category-Level and Brand-Level Purchases Using Potential Outcome Approach With Bayes Inference

Kei Miyazaki, Takahiro Hoshino*, Ulf Böckenholt

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

We propose an econometric two-stage model for category-level purchase and brand-level purchase that allows for simultaneous brand purchases in the analysis of scanner panel data. The proposed model formulation is consistent with the traditional theory of consumer behavior. We conduct Bayesian estimation with the Markov chain Monte Carlo algorithm for our proposed model. The simulation studies show that previously proposed related models can cause severe bias in predicting future brand choices, while the proposed method can effectively predict them. Additionally in a marketing application, the proposed method can examine brand switching behaviors that existing methods cannot. Moreover, we show that the prediction accuracy of the proposed method is higher than that of existing methods.

Original languageEnglish (US)
Pages (from-to)622-635
Number of pages14
JournalJournal of Business and Economic Statistics
Volume39
Issue number3
DOIs
StatePublished - 2021

Funding

This work was supported by JSPS KAKENHI Grant Numbers JP18H03209, JP16H06323, JP16H02013, JP26285151 for T. Hoshino. We would like to express our sincere thanks to the associate editor and reviewers for their valuable advice and comments. The computational results are obtained using Ox version 6.0 (Doornik 2009).

Keywords

  • Bayes estimation
  • Missing data
  • Multiple-brand purchase
  • Potential outcomes
  • Product-category purchase

ASJC Scopus subject areas

  • Statistics and Probability
  • Social Sciences (miscellaneous)
  • Economics and Econometrics
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'Dynamic Two Stage Modeling for Category-Level and Brand-Level Purchases Using Potential Outcome Approach With Bayes Inference'. Together they form a unique fingerprint.

Cite this