John E. Walker Department hosts workshop on modern econometric methods

This spring, the John E. Walker Department of Economics at the Wilbur O. and Ann Powers College of Business organized and hosted a two-day workshop on empirical methods in social science research. This workshop brought together some of the most highly cited applied statisticians and econometricians in the world. Their research, focused on causal inference, has helped to reshape the way social scientists analyze data.

The workshop focused on frontier techniques that define the ongoing “causal revolution” in empirical research. The event was specially designed for faculty members to gain exposure to the latest applied methods in social science and see some of their real-world applications.

Scott Baier, acting associate dean of research, says, “Hosting this Workshop on Empirical Methods in the Social Sciences at Clemson University provided a unique opportunity to bring frontier methodological advances directly to our research community. With presentations from some of the most highly cited contributors to the “causal revolution in social sciences”, the faculty and graduate students received exposure to cutting-edge methods that will greatly enhance our capacity to produce policy-relevant research grounded in credible causal inference.”

The following researchers presented at the workshop, with topics ranging from staggered treatment designs to the subtleties of bunching in treatment effects.

Paul Wilson, Clemson University

Paul Wilson, J. Wilson Newman Professor of Economics and Professor of Computer Science, presented the use of linear probability models in his paper “Zombie Econometrics: The Linear Probability Specification.” Wilson emphasized the pitfalls of applying these models without accounting for their inherent limitations, urging researchers to exercise care in modeling binary outcomes.

Pedro Sant’Anna, Emory University

Pedro Sant’Anna, an associate professor of economics at Emory University and an Amazon Visiting Academic. Sant’Anna has published in top journals in economics and has given guest lectures and seminars on DiD topics at leading universities around the world. In his presentation, he shared a checklist for applying DiD in settings with treatment effect heterogeneity and staggered adoption

Jeffery Wooldridge, Michigan State University

Jeffrey Wooldridge, University Distinguished Professor of Economics, provided an overview of regression-based difference-in-difference estimators in staggered designs, possibly also with staggered exit. With nearly 100 refereed publications and over 160,000 Google Scholar citations, Wooldridge’s talk on Regression-Based Difference-in-Differences Designs with Staggered Entry offered a unified and intuitive framework.

Matias Cattaneo, Princeton University

Matias Cattaneo, professor of operations research and financial engineering, has research that spans econometrics, statistics, data science and decision science, with applications to program evaluation and causal inference. Cattaneo discussed advancements in Regression Discontinuity Designs (RDD). He showed how the standard RDD framework can yield biased estimates when applied to spatial data and offered a straightforward correction method.

Colin Cameron, University of California-Davis

Colin Cameron, Distinguished Professor of Economics, presented the challenges of inference in the presence of spatial correlation. He warned that failing to account for geographic dependence in data can result in underestimated standard errors and overconfident statistical conclusions.

Brantly Callaway, University of Georgia

Brantly Callaway, associate professor, has research focused on how to use panel data (repeated observations on the same individual or firm over time) to think about the causal effects of economic policies. At the workshop, he discussed addressing DiD designs with continuous treatments, an increasingly common scenario in policy research.

Carolina Caetano, University of Georgia

Carolina Caetano, an assistant professor of economics, introduced strategies for identifying treatment effects in the absence of clean randomization. Her presentation on kinks, thresholds and bunching designs helped researchers think more clearly about the messy realities of observational data.

About the John E. Walker Department of Economics

At Clemson, the economics coursework lays a robust foundation for successful careers in business, law and government. The economics curriculum also prepares our students for graduate studies in economics, law or related disciplines. Several educational paths are offered through the department, including bachelor of arts and science degrees, master’s in economics and a Ph.D. in economics.