Nonparametric Identification and Estimation of the Extended Roy Model

46 Pages Posted: 18 Oct 2022

See all articles by Ji Hyung Lee

Ji Hyung Lee

University of Illinois at Urbana-Champaign - Department of Economics

Byoung Park

University at Albany (SUNY)

Date Written: October 8, 2022

Abstract

We propose a new identification and estimation method for the extended Roy model, in which the agents maximize their utility rather than just their outcome. We nonparametrically identify the joint distribution of potential outcomes, which is of great importance particularly in treatment effect analysis. The identification is achieved by matching the indifferent agents across choices, who are identified by the local instrumental variable method. We exploit the extended Roy model structure and the monotonicity assumption but do not require any functional form assumption nor any support assumption. Based on the identification result, we propose a nonparametric estimation procedure that builds upon a simulation-based method proposed by Dette et al. (2006). The estimator is easy to implement and possesses a standard nonparametric rate of convergence. The estimator's efficacy in finite samples is examined in Monte Carlo simulations. An empirical illustration on Malawian farmers' hybrid maize adoption is provided.

Keywords: self-selection, Roy model, nonseparable model, nonparametric identification, treatment effect, simulation-based estimator

JEL Classification: C14, C35, C36, C51, C53

Suggested Citation

Lee, Ji Hyung and Park, Byoung, Nonparametric Identification and Estimation of the Extended Roy Model (October 8, 2022). Journal of Econometrics, Forthcoming, Available at SSRN: https://ssrn.com/abstract=4242294

Ji Hyung Lee

University of Illinois at Urbana-Champaign - Department of Economics ( email )

214 David Kinley Hall
1407 W. Gregory
Urbana, IL 61801
United States

Byoung Park (Contact Author)

University at Albany (SUNY)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
50
Abstract Views
282
PlumX Metrics