Yogurts Choose Consumers? Estimation of Random Utility Models via Two-Sided Matching
54 Pages Posted: 9 Mar 2017 Last revised: 4 Oct 2018
Date Written: May 16, 2018
In this paper we show that the problem of demand inversion in multinomial choice models is equivalent to the determination of stable outcomes in matching models. This result is very general and applies to random utility models that are not necessarily additive or smooth. Based on this equivalence, we argue that the algorithms for the determination of stable matchings can provide effective computational methods to inverse multinomial choice models, and we give a numerical benchmark of these algorithms. Our approach allows to estimate models that were previously difficult to estimate, such as the pure characteristics model, as well as nonadditive random utility models. The equivalence also allows to exploit the theory of stable matchings in order to describe important properties of the set of utilities solution to the demand inversion problem, and to study the cases of existence and uniqueness of identified utilities, as well as obtain consistency results.
Keywords: Random Utility Models, Two-Sided Matching, Optimal Transport, Pure Characteristics Model, Deferred Acceptance, Partial Identification
JEL Classification: C51, C60
Suggested Citation: Suggested Citation