An empirical likelihood estimator for average treatment effects in stratified randomized designs

• Grant number: 11/02592-7
• Support Opportunities: Scholarships abroad - Research
• Start date: August 01, 2011
• End date: March 13, 2012
• Field of knowledge: Applied Social Sciences - Economics - Economics of Human Resources
• Principal Investigator: Sergio Pinheiro Firpo
• Grantee: Sergio Pinheiro Firpo
• Host Investigator: Richard Blundell
• Host Institution: Escola de Economia de São Paulo (EESP). Fundação Getúlio Vargas (FGV). São Paulo , SP, Brazil
• Institution abroad: University College London (UCL), England

Abstract

This project proposes a new methodology for both testing for imbalances and estimating the average treatment e¤ect parameter (ATE) controlling for the bias that might arise if within strata randomization was not properly performed. In that sense, our results are expected to facilitate the task of applied researchers that have become more involved with research designs. Finally, a great advantage of our method is to allow comparabilityacross results, since no arbitrary choices are required from users of the methodology, suchas deciding whether to drop or not entire strata that seem to be imbalanced in covariates. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

FIRPO, SERGIO; FOGUEL, MIGUEL N.; JALES, HUGO. Balancing tests in stratified randomized controlled trials: A cautionary note. ECONOMICS LETTERS, v. 186, JAN 2020. (11/02592-7)