Mixture Choice Data: Revealing Preferences and Cognition

Abstract

Mixture choice data consist of the joint distribution of choices of a group of agents from a collection of menus, comprising the implied stochastic choice function plus any cross-menu correlations. When agents are heterogeneous with respect to both preferences and other aspects of cognition, we show that these two determinants of behavior are identified simultaneously by suitable mixture choice data. We also demonstrate how this finding can be extended to allow for specialized assumptions about cognition, focusing on models of random satisficing thresholds and "quantal Fechnerian" choice.

Correlated choice

“…In addition, this type of data has been used to study the return on education by studying twins (see Ashenfelter and Krueger (1994) and Miller, Mulvey, and Martin (1995)). 3 Another recent paper, Dardanoni, Manzini, Mariotti, Petri, and Tyson (2022), studies a richer form of data. Dardanoni et al (2022) studies probability distributions over choice functions.…”

“…33 Another recent paper, Dardanoni, Manzini, Mariotti, Petri, and Tyson (2022), studies a richer form of data. Dardanoni et al (2022) studies probability distributions over choice functions. Our data considers probability distributions over alternative pairs from product sets. …”

Presidential Address: Economics and Measurement: New Measures to Model Decision Making

“…But in real life, preferences are not directly observable: we have to discover peoples' preferences from observing their behavior." (Chapter 7).2 Berry, Levinsohn, and Pakes (2004) is an exception, where rankings or 'second hypothetical choices' are used in the identification of demand models for cars.3 A recent paper byDardanoni, Manzini, Mariotti, Petri, and Tyson (2022) discusses the restrictions that the use of choice data requires for models with preference and other types of heterogeneity.…”

“…33 A recent paper by Dardanoni, Manzini, Mariotti, Petri, and Tyson (2022) discusses the restrictions that the use of choice data requires for models with preference and other types of heterogeneity. …”