Salience biased stochastic choice

• Grant number: 20/11460-6
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: June 01, 2021
• End date: December 31, 2023
• Field of knowledge: Applied Social Sciences - Economics - Economic Theory
• Principal Investigator: José Heleno Faro
• Grantee: Antonio Daniel Ricardo Engracia Caluz
• Host Institution: Instituto de Ensino e Pesquisa (Insper). São Paulo , SP, Brazil

Abstract

Salience bias behavior has become one of the central topics in the recent decision-making economic literature. There is both applied and theoretical study documenting that the choices can be distorted in favor of salient payoffs, attributes, or goods, different from what the predicted in the traditional rational behavior. Salience Bias can arise in a simplification in the decision-making in which there is some inattention by individuals. This project proposes to model the salience bias behavior introducing this phenomenon in a way not done so far, in stochastic choice models. In this way, we can match the salience theory and real data, which is noisy (MANZINI; MARIOTTI, 2014). We aim to characterize salience bias through three different classes of stochastic choice rules: A nested-logit type model with salience bias; a Luce (1959) model with salience; and an elimination model a la Tversky (1972). For each stochastic choice specification, we intend to provide the corresponding behavioral foundation, which can elucidate its testable implications. Moreover, an important goal is to develop an econometric analysis of qualitative response models where the endogenous variables may result from economic behavior that is intrinsically categorical and salience biased. (AU)