Faculty

Professor Hwang to publish “A Doubly Corrected Robust Variance Estimator for Linear GMM” in the Journal of Econometrics

Posted on by

Professor Jungbin Hwang has had his article A Doubly Corrected Robust Variance Estimator for Linear GMM accepted for publication in the Journal of Econometrics, one of the top scholarly journals in theoretical econometrics.

The paper proposes a new finite sample corrected variance estimator for the linear generalized method of moments (GMM) including the one-step, two-step, and iterated estimators. The formula additionally corrects for the over-identification bias in variance estimation on top of the commonly used finite sample correction of Windmeijer (2005) which corrects for the bias from estimating the efficient weight matrix, so is doubly corrected. An important feature of the proposed double correction is that it automatically provides robustness to misspecification of the moment condition. In contrast, the conventional variance estimator and the Windmeijer correction are inconsistent under misspecification. That is, the proposed double correction formula provides a convenient way to obtain improved inference under correct specification and robustness against misspecification at the same time.

This article, authored with with Seojeong Lee (UNSW) and Byunghoon Kang (Lancaster Univ), is one of two that Professor Hwang has recently had accepted for publication. Details of the other article may be found at:

Professor Jungbin Hwang publishes “Simple and Trustworthy Cluster-Robust GMM Inference” in the Journal of Econometrics

Professor Jungbin Hwang publishes “Simple and Trustworthy Cluster-Robust GMM Inference” in the Journal of Econometrics

Posted on by

Professor Jungbin Hwang has had his article Simple and Trustworthy Cluster-Robust GMM Inference accepted for publication in the Journal of Econometrics, a top field-journal in econometrics.

This paper develops a new asymptotic theory for GMM estimation and inference in the presence of clustered dependence. The key feature of the alternative asymptotic is that the number of clusters is regarded as “fixed” as the sample size increases. The paper shows that the Wald and t-tests in two-step GMM are asymptotically pivotal only if one recenters the estimated moment process in the clustered covariance estimator (CCE). Also, the J statistic, the trinity of two-step GMM statistics (QLR, LM, and Wald), and the t statistic can be modified to have an asymptotic standard F distribution or t distribution.

The paper also first suggests a finite-sample variance correction in the literature of cluster-robust methods and further improves the F and t approximations’ accuracy. The proposed tests in this paper are very appealing to practitioners because the test statistics are simple modifications of conventional GMM test statistics, and critical values are readily available from F and t tables without any extra simulations or resampling steps.

This sole authored article is one of two that Professor Hwang has recently had accepted for publication. Details of the other article may be found at:

Professor Hwang to publish “A Doubly Corrected Robust Variance Estimator for Linear GMM” in the Journal of Econometrics

Professor Couch Publishes in Journal of Public Economics

Posted on by

Professor Kenneth Couch and his co-authors, Robert Fairlie (California Santa Cruz) and Huanan Xu (Indiana University South Bend) have published an article “Early Evidence of the Impact of COVID-19 on Minority Employment” In the Journal of Public Economics.

Article Abstract:

This paper provides early evidence of the impacts of the COVID-19 pandemic on minority unemployment in the United States. In the first month following March adoptions of social distancing measures by states, unemployment rose to 14.5 percent but a much higher 24.4 percent when we correct for potential data misclassification noted by the BLS. Using the official definition, unemployment in April 2020 among African-Americans rose by less than what would have been anticipated (to 16.6 percent) based on previous recessions, and the long-term ordering of unemployment across racial/ethnic groups was altered with Latinx unemployment (18.2 percent) rising for the first time to the highest among major groups. Difference-in-difference estimates confirm that the initial gap in unemployment between whites and blacks in April was not different than in periods prior to the pandemic; however, the racial gap expanded as unemployment for whites declined in the next two months but was largely stagnant for blacks. The initially large gap in unemployment between whites and Latinx in April was sustained in May and June as unemployment declined similarly for both groups. Non-linear decompositions show a favorable industry distribution partly protected black employment during the early stages of the pandemic, but that an unfavorable occupational distribution and lower average skills levels placed them at higher risk of job losses. An unfavorable occupational distribution and lower skills contributed to a sharply widened Latinx-white unemployment gap that moderated over time as rehiring occurred. These findings of disproportionate impacts on minority unemployment raise important concerns regarding lost earnings and wealth, and longer-term consequences of the pandemic on racial inequality in the United States.

The early release version of this article can be found at this link:

https://www.sciencedirect.com/science/article/pii/S0047272720301511

Professor Agüero publishes in Economic Development and Cultural Change

Posted on by

Professor Jorge Agüero’s paper “Measuring Violence Against Women with Experimental Methods” has been accepted for publication by Economic Development and Cultural Change

The working paper may be found online at: https://ideas.repec.org/p/uct/uconnp/2020-14.html

Title: Measuring Violence Against Women with Experimental Methods
Authors: Jorge Aguero and Veronica Frisancho

Abstract: The prevalence of intimate partner violence is a central indicator of the Sustainable Development Goals for women’s agency. However, measuring this indicator largely relies on self-reports that could suffer from severe misreporting if women face high costs of revealing their victim status. We study the degree of misreporting in surveys that have been identified as the best source of data, such as the widely used Demographic and Health Surveys (DHS). Focusing on a sample of women in impoverished urban areas of Lima, Peru, we conduct an experiment that replicates direct measures from these surveys and compares them against list experiments, a method that provides greater privacy to respondents. We find no significant differences across direct and indirect methods in any of the seven acts of physical and sexual violence considered. This result largely persists when testing across sixteen different subgroups and accounting for multiple hypothesis testing.

Pilot Project Approved for Professor Agüero and PhD Student Mendiola

Posted on by

Professor Jorge Agüero and third year PhD student Miranda Mendiola’s proposal “Role models: Information and Gender Stereotypes” for a pilot project, sponsored by the Innovation Laboratory for Cost-Effective Educational Policy – MineduLAB in the Peruvian Ministry of Education, has been approved.

Their project has the objective of reducing gender stereotypes and improving grades for high school students through the use of role models. Traditionally, efforts to reduce gender gaps have focused on empowering women. Professor Agüero and Miranda’s project focuses on changing the perception of both genders’ abilities by showing students movies that have young main characters being successful in careers that are nontraditional for their gender. They hope to improve women’s scores in STEM courses, where they traditionally perform worse, and also to improve men’s scores in courses they traditionally struggle with (Spanish and history). They will measure changes in gender bias through a questionnaire and a game, with the objective of measuring both explicit and implicit biases. 

This project will hopefully be a pilot for a larger project in Peru, aiding in the reduction of gender bias in Peruvian schools.

Professor Ray publishes in Empirical Economics

Posted on by

Professor Subhash Ray published his recent paper “Unrestricted geometric distance functions and the Geometric Young productivity index: an analysis of Indian manufacturing” coauthored with Arnab Deb (Associate Professor, International Management Institute New Delhi) and Kankana Mukherjee (Associate Professor, Babson College) in Empirical Economics.

At this point, the paper is available online at https://doi.org/10.1007/s00181-020-01925-0.

Both of his coauthors are his former PhD students: Arnab Deb (PhD UConn 2012) and Kankana Mukherjee (PhD UConn 1997).

Professors Harmon and Tomolonis Publish in Journal of Economic Education

Posted on by

JEE LogoOskar Harmon and Paul Tomolonis (UConn PhD 2017) have co-authored the article “Learning Tableau – A data visualization tool”, published in the Journal of Economic Education.

ABSTRACT: “Doing economics” is an important theme of undergraduate economics programs. Capstone courses increasingly include instruction in “data literacy” and the STEM-related skills of quantitative and empirical methods. Because the professional discipline has moved in this direction and because of greater employer demand for these skills, data visualization is a key component of data literacy. Tableau is a free data visualization software widely used in the data analytics industry. In this article, the authors introduce an exercise that teaches the fundamental Tableau concepts and commands needed to create charts, assemble them in a dashboard, and tell a story of patterns observed in the data. The exercise assumes no prior experience in Tableau and is appropriate for undergraduate upper-level economics courses or an empirical methods course.

The article is available at the JEE website

NBER Releases Research on COVID-19 Impacts on Minority Unemployment Co-authored by UConn Economics Faculty

Posted on by

A working paper has been released by the National Bureau of Economic Research that examines the impact of COVID-19 on minority unemployment through the most recent release of Current Population Survey (CPS) data for April of 2020.

The research finds that unemployment of blacks (at 16.6 percent) has not been impacted as severely as during past downturns although their unemployment rate is above the national average of 14.7 percent.  In comparison, Latinx unemployment (at 18.2 percent) has been much more impacted than in recent months or the Great Recession.  Historically, unemployment of blacks would be greater than that of Latinx throughout the business cycle.  In the April CPS data, for the first time, unemployment of Latinx is higher.  The analysis reveals that the disproportionate impact among the Latinx is related to lower levels of education, less work experience, and a concentration of employment in industries and occupations that left them more vulnerable to job loss.

The research is co-authored by Robert Fairlie of the University of California Santa Cruz, UConn Faculty member Ken Couch, and Huanan Xu of Indiana University South Bend.  Xu is an alumni of the UConn Ph.D. program in economics.

The working paper can be found at this link:

https://www.nber.org/papers/w27246.pdf