I apply and develop techniques for evaluating the effects of treatments, policies, and ads through the lens of Donald Rubin’s causal model, as described by Paul Holland. Particular methodological interests include matching, missing data, measurement error, covariate balance, covariance adjustment, randomization tests, sensitivity analysis, cluster-robust variance estimation, and effect generalizability. I have applied my methods to the Texas Education Agency’s Additional Days School Year (ADSY) initiative and the SUNFISH trial for assessing the effects of risdiplam on spinal muscular atrophy. Previous work includes ad testing and election forecasting at Civis Analytics. I am currently a fifth-year Ph.D. candidate in statistics at the University of Michigan advised by Professor Ben Hansen.
Publications
Under Review
- Wasserman, J, Pang, H., Zhu, J. Addressing missing data in clinical trials with a hybrid control arm. Under review at Statistics in Medicine.
- Wasserman, J, Elliott, M.R., Hansen, B.B. Propensity score adjustment when errors in achievement measures inform treatment assignment. Under review at Journal of Educational and Behavioral Statistics.
In preparation
- Wasserman, J, Hansen, B.B. Two-stage estimation for flexible covariance adjustment of treatment effect estimates.
Software
Research Leadership
- Project leader (Undergraduate Research Program in Statistics, Winter 2024, University of Michigan)
- Undergraduate researchers built hierarchical linear models for predicting student standardized test scores
- Co-mentor (Big Data Summer Institute, Summer 2023, University of Michigan)
- Undergraduate researchers used data linkage techniques to predict tumor severity from imaging and clinical data
Talks
- Propensity scores for coarsened data due to small-cell suppression of subgroup covariates: The case of school matching in a typical U.S. state (65th International Statistics Institute World Statistics Congress, October 2025, The Hague, Netherlands; slides)
propertee: Flexible covariance adjustment and improved standard errors in analyses of intact clusters (useR! Conference, August 2025, Durham, NC; slides)- Separating covariance adjustment from causal effect estimation to leverage auxiliary datasets (Association for Education, Finance, and Policy 49th Annual Conference, March 2024, Baltimore, MD; slides)
- Covariate adjustment with non-experimental units using
propertee(Joint Statistical Meetings, August 2023, Toronto, ON; poster) - Covariate adjustment with non-experimental units using
propertee(American Causal Inference Conference, May 2023, Austin, TX; poster)
Teaching
- Advanced Regression Analysis
- Introduction to the Design of Experiments
- Undergraduate Statistics/Data Science Capstone Seminar
- Survey Sampling Techniques
- Introduction to Statistics and Data Analysis