SELECTED RESEARCH

Title: Consumer Theory with Non-Parametric Taste Uncertainty and Individual Heterogeneity
Authors: Christopher Dobronyi, and Christian Gouriéroux
Conferences: ESEMEconometric Society European Meeting (2021), CFE (2020)Conference on Computational and Financial Econometrics (2020)
Seminars: Ottawa, Toronto, WaterlooUniversity of Ottawa, University of Toronto, University of Waterloo
More: Slides

This paper introduces two models of non-parametric random utility for demand systems. In each model, individual-level heterogeneity is characterized by a distribution $\pi\in\Pi$ of taste parameters, and heterogeneity across consumers is introduced using a distribution $F$ over the distributions in $\Pi$. Demand is non-separable and heterogeneity is infinite-dimensional. Two frameworks for estimation are considered: a Bayesian framework, and a hyperparametric framework.

Title: Stochastic Revealed Preference: A Non-Parametric Approach for Scanner Data
Authors: Christopher Dobronyi, and Christian Gouriéroux
Conferences: ESWCEconometric Society World Congress (2020), $\textbf{EC}^2$ Conference on High Dimensional Modeling in Time Series (2020)
Seminars: TorontoUniversity of Toronto
More: Slides, Recorded ESWC Presentation

This paper uses random fields to develop a non-parametric model of consumption with infinite-dimensional heterogeneity appropriate for scanner data. Consumers are allowed to be quasi-rational. The theory of generalized functions is used to test the integrability of expected demand at a parametric rate. When variation in preferences is small, they can be recovered by approximating the relationship between preferences and demand using a first-order expansion and applying an analogue of the delta-method. Else, preferences can be recovered numerically.

Title: Moment Equalities and Inequalities in Dynamic Panel Logit Models with Fixed EffectsFEs
Authors: Christopher Dobronyi, Jiaying Gu, and Kyoo il Kim

This paper uses moment problems to characterize the sharp identified set for the parameters in a class of dynamic panel logit models with fixed effects. All information is characterized by a small collection of moment equalities and inequalities. This result is used to identify sharp bounds for the parameters in models without point identification, rule out false roots in models that cannot be identified using only moment equalities, and identify sharp bounds on functionals of the distribution of heterogeneity.

Title: Goal Setting, Academic Reminders, and College Success: A Large-Scale Field Experiment
Authors: Christopher Dobronyi, Philip Oreopoulos, and Uros Petronijevic
Journal Information: Journal of Research on Educational EffectivenessJREE 12(1): 38-66

This paper presents an independent large-scale experimental evaluation of two online goal-setting interventions. Approximately 1,400 first-year undergraduate students at the University of Toronto were randomly assigned into either treatment or control. Half of the treated students were also offered the opportunity to receive academic reminders. This paper finds no evidence of an effect on grade point average, course credits, or second-year persistence. These results hold by subsample, and across a large number of specifications.

TEACHING

Course: Perspectives on Computational Modeling for Economics (ECMA 31140 / MACS 30150)
Institution: University of Chicago
Level: Graduate

This course introduces graduate students to computational problems in microeconomics and econometrics. We will cover a wide variety of topics related to convex optimization, non-parametric estimation, revealed preference, demand, discrete choice, matching, and game theory.