Hi, I’m Walter Dempsey 👋
I am an Associate Professor of Biostatistics and an Associate Research Professor in the d3lab located in the Institute of Social Research at the University of Michigan. My research focuses on Statistical Methods for Digital and Mobile Health. I list my statistical papers below. A full publication list can be found on Google Scholar.
Education
-
University of Chicago
Department of Statistics
Ph.D.
-
University of Michigan
Department of Statistics
Postdoctoral Research Fellow
-
Harvard University
Department of Statistics
Postdoctoral Research Fellow
I list my statistical papers below. * indicates mentored student or postdoctoral fellow. A full publication list can be found on
Posts about Methods, Software, Tutorials, Lab Updates, and (sometimes) My Opinions/Side Projects.
Most recent post:
Evaluating Time-Varying Treatment Effects in Hybrid SMART-MRT Designs
methodsA statistical framework for jointly estimating synergistic and marginal causal effects when digital and human-delivered interventions operate at multiple timescales.
A list of recent courses that I have taught at the University of Michigan.
BIOSTAT617
01/01/2019 - 06/30/2023Theory and Methods in Survey Design. Theory underlying sample designs and estimation procedures commonly used in survey practice. The latest version of the course can be found on Canvas
BIOSTAT629
08/30/2019 - 08/30/2020Case Studies In Health Big Data. A project-based course that integrates all competencies learned in the Health Data Science Curriculum to provide a culminating research experience. Students work on two to three health big data projects, through which they learn to identify scientific objectives and analytical strategies and report findings through oral presentation and written documents. The latest version of the course can be found on Canvas
A list of recent graduate students and postdoctoral research fellows.
Yaxuan Huang
PhD Student in Biostatistics. Yaxuan's research focuses on developing reinforcement learning methods for decision-making with missing data. She is also interested in real-world deployment of these methods in online mobile health studies.
Rachel Tucker Gonzalez
PhD Student in Biostatistics (with Phil Boonstra). Rachel's research interests include Bayesian nonparametric methods for multistate models, with applications to win ratio estimation. She is a member of the Cancer Biostatistics Training Program at the University of Michigan and works on a variety of applied projects across cancer sites.
Yuxuan Chen
PhD Student in Biostatistics. Works on selective inference to ensure valid statistical inference for effect moderation when models are chosen in a data-driven manner using micro-randomized trial data.
Tzu-Hsuan Lin
PhD Student in Biostatistics.
Youqi Yang
PhD Student in Biostatistics (with Bhramar Mukherjee). Research focuses on statistical methods for causal inference in mediation within micro-randomized trials, as well as treatment heterogeneity within randomized controlled trials.
Swaraj Bose
PhD Student in Biostatistics. Research focuses on building prediction intervals for Individual Treatment Effects (ITEs) for Micro-randomized trials (MRTs) using conformal inference techniques for decision support.
Madeline Abbott
Graduated, 2024. Research Associate in the Department of Biostatistics at Harvard University. F31 Grant Recipient, Best Presentation at MSSISS 2023. Worked on statistical methods for intensive longitudinal data arising from mobile health trials, specifically latent variable models and joint models with recurrent event processes.
Easton Huch
Graduated, 2025. Postdoctoral Fellow in Quantitative Marketing at Johns Hopkins University. Worked on robust Bayesian methods for causal inference with applications to time-varying effect estimation in mobile health studies.
Hera Shi
Graduated, 2023. Postdoctoral Research Fellow at the University of Cambridge with Qingyuan Zhao. Worked on statistical methods for time-varying treatment effects in the context of micro-randomized trials.