Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 43 - Advances in Modeling Consumer Choice Behavior
Type: Invited
Date/Time: Sunday, August 7, 2022 : 4:00 PM to 5:50 PM
Sponsor: Business and Economic Statistics Section
Abstract #320670
Title: Consumer Choice with Dependent Consideration Sets
Author(s): Maura Coughlin* and Matthew Thirkettle
Companies: Rice University and Rice University
Keywords: limited consideration; discrete choice; identification; econometrics
Abstract:

We develop a discrete choice model that incorporates unobserved heterogeneity in consideration sets. We depart from the standard approach that assumes consideration of each alternative is independent of all other alternatives (e.g., Manski, 1977; Goeree 2008; Manzini and Mariotti, 2014; Abaluck and Adams-Prassl 2020). Instead, we allow for non-parametric dependence in the formation of consideration sets between alternatives belonging to a given category, while still maintaining independence across categories. Our method accommodates cases where consideration of alternatives can depend on unobservable factors that result in certain alternatives jointly appearing in consideration sets more often. For example, this allows for the case where a consumer purchases a product through an agent, such as insurance, and the agent presents the alternatives to the consumer in an unobserved (to the researcher) order that yields correlation between which plans are considered and eventually selected. We show how to identify the consideration set process as well as preferences. We then validate our approach through a simulation study.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program