Center for Causal Inference Symposium

Date:	Thursday, August 18, 2022
Time:	10:30 a.m. - 6:00 p.m. EDT/7:30 a.m. - 3 p.m. PDT

Overview

At the heart of policy analysis is cause and effect: How can we determine or quantify the actual impact of an action or policy? Without proper methods, recklessly inferring cause and effect from observed relationships can inform unhelpful policies at best, and dangerous ones at worst. More than ever, the research community needs the development and understanding of causal inference methods.

Following the success of the inaugural 2021 event, we were pleased to host the 2022 RAND Center for Causal Inference (CCI) Symposium. This one-day interdisciplinary virtual symposium offered researchers the opportunity to present and learn about cutting-edge causal inference research in statistics, econometrics, and other quantitative fields.

Although the symposium took place online, it was not recorded.

Program

The symposium included four sessions, each featuring four speakers and a brief, moderated discussion.

Presenting authors are listed in italics.

10:30 a.m.

Opening Remarks

• Maria DeYoreo

10:35 a.m. - 12:10 p.m.

Session 1: “Big Picture”

Nonparametric and Equitable Quality Measurement

• Denis Agniel, Irineo Cabreros, Marc Elliott, Cheryl Damberg, Rhianna Rogers

A Semiparametric Method for Evaluating Causal Effects in the Presence of Error-Prone Covariates

• Jianxuan Liu and Wei Li

Elements of External Validity: Framework, Design, and Analysis

• Naoki Egami and Erin Hartman

What Could We Have Safely Learned About the Effects of Dexamethasone, Remdesivir, and Hydroxychloroquine on COVID-19 Mortality Before RCTs?

• Chad Hazlett and Ami Wulf

Moderated Discussion of Session 1

• Maria DeYoreo

12:10 - 12:30 p.m.

Break

12:30 - 2:05 p.m.

Session 2: “Quasi-Experimental Methods”

Bayesian and Frequentist Inference for Synthetic Controls

• Jaume Vives-i-Bastida and Ignacio Martinez

Statistical Inference for the Factor Model Approach to Estimate Causal Effects in Quasi-Experimental Settings

• Kathleen Li and Garrett Sonnier

Optimal Linear Instrumental Variables Approximations

• Wei Li and Juan Carlos Escanciano

Two Stage Curvature Identification with Machine Learning: Causal Inference with Possibly Invalid Instrumental Variables

• Zijian Guo and Peter Buhlmann

Moderated Discussion of Session 2

• Jessie Coe

2:05 - 2:25 p.m.

Break

2:25 - 4:00 p.m.

Session 3: “Tools and Applications”

If You Build it, Will They Vaccinate? The Impact of COVID-19 Vaccine Sites on Vaccination Rates and Outcomes

• John S. Brownstein, Jonathan Cantor, Benjamin Rader, Kosali I. Simon, and Christopher Whaley

A Tree-based Model Averaging Approach for Personalized Treatment Effect Estimation from Heterogeneous Data Sources

• Xiaoqing Tan, Chung-Chou H. Chang, Ling Zhou, and Lu Tang

Image-based Treatment Effect Heterogeneity

• Connor T. Jerzak, Fredrik Johansson, and Adel Daoud

Power under multiplicity project (PUMP): Estimating power when adjusting for multiple outcomes in multilevel experiments

• Kristen Hunter, Luke Miratrix, Kristin Porter, and Zarni Htet

Moderated Discussion of Session 3

• John Engberg

4:00 - 4:20 p.m.

Break

4:20 - 5:55 p.m.

Session 4: “Balance and Weighting”

Targeted Learning in Observational Studies with Multilevel Treatments: An Evaluation of Antipsychotic Drug Treatment Safety

• Jason Poulos, Marcela Horvitz-Lennon, Katya Zelevinsky, Thomas Huijskens, Pooja Tyagi, Jiaju Yan, Jordi Diaz, Tudor Cristea-Platon, Sharon-Lise Normand

Variance-based Sensitivity Analysis for Weighted Estimators for More Informative Bounds

• Melody Huang and Sam Pimentel

Outcome Adaptive Variable Selection Procedure for the Covariate Balancing Propensity Score

• Mingyuan Zhang, Yiming Xu, Jonathan Chipman, Matthew H. Samore, and Yue Zhang

Reducing Bias in Difference-in-Differences Models Using Entropy Balancing

• Matt Cefalu, Brian Vegetabile, Federico Girosi, Christine Eibner, and Michael Dworsky

Moderated Discussion of Session 4

• Beth Ann Griffin

5:55 - 6:00 p.m.

Closing

• Beth Ann Griffin

Presenters' and Moderators' Bios

Denis Agniel is a statistician at RAND.

Jessie Coe (she/her) is an associate economist at RAND and professor of policy analysis at the Pardee RAND Graduate School.

Maria DeYoreo is a statistician and codirector of the Center for Causal Inference at RAND.

Michael Dworsky is a senior economist at RAND.

John Engberg is a senior economist at RAND.

Beth Ann Griffin is a senior statistician and acting codirector of the Center for Causal Inference at RAND.

Zijian Guo is an assistant professor at Rutgers University.

Erin Hartman is an assistant professor at University of California, Berkeley.

Chad Hazlett is an associate professor in the departments of Statistics and Political Science at UCLA.

Melody Huang is a Ph.D. candidate at University of California, Berkeley.

Kristen Hunter is a Ph.D. candidate at Harvard University.

Connor T. Jerzak is an assistant professor at the University of Texas at Austin.

Kathleen Li is an assistant professor of marketing at the University of Texas at Austin.

Wei Li is an assistant professor at Syracuse University.

Jianxuan Liu is an assistant professor at Syracuse University.

Jason Poulos is a postdoctoral fellow at Harvard Medical School.

Xiaoqing "Ellen" Tan is a Ph.D. candidate at the University of Pittsburgh.

Jaume Vives-i-Bastida is a Ph.D. candidate at MIT.

Christopher Whaley is an economist at RAND and a professor of policy analysis at the Pardee RAND Graduate School.

Mingyuan Zhang is a graduate student in biostatistics at the University of Utah.

Questions?

For more information about CCI, visit our website or email us at causalinference@rand.org