Keith Barnatchez

Hi! I’m a Postdoctoral Researcher in the biostatistics department at Johns Hopkins University. My research focuses on developing methods for causal inference when data is collected from multiple sources subject to different potential biases (e.g. hospital networks). My methods frequently use tools from machine learning and semiparametric theory to reduce reliance on restrictive modeling assumptions.

Before joining Hopkins, I spent three years working as a research assistant at the Federal Reserve Bank of Boston, and I completed a PhD in biostatistics at Harvard University,

You can contact me at kbarnat1@jh.edu.

News

• Our paper Debiased Machine Learning for Conformal Prediction of Counterfactual Outcomes Under Runtime Confounding was selected for an oral presentation at the 5th Conference on Causal Learning and Reasoning (CLeaR)!

• Our paper Efficient Estimation of Causal Effects Under Two-Phase Sampling with Error-Prone Outcome and Treatment Measurements was co-awarded best student poster at the 2025 New England Rare Disease Statistics (NERDS) Workshop! Our pre-print can be found on arXiv.

• Our manuscript Flexible and Efficient Estimation of Causal Effects with Error-Prone Exposures: A Control Variates Approach for Measurement Error was accepted to Biometrics! You can view our pre-print on arXiv.

• Our R package drcmd: Doubly-Robust Causal Inference with Missing Data is now available through GitHub. Please check back frequently, as the packge is being actively updated!