Machine Learning for Causal Inference

Brown Bag Lunch – Please RSVP.

Machine learning provides exciting opportunities to better understand risk factors, to build improved prediction algorithms, and to examine the causal relationships between variables. Still, there are many sources of association, including direct effects, indirect effects, measured confounding, unmeasured confounding, and selection bias. Methods to delineate causation from correlation are perhaps more pressing now than ever. As a concrete example, we focus on assessing disease burden and control in the SEARCH Study, a large (>320,000-persons) community randomized trial for HIV prevention and treatment in rural Kenya and Uganda (NCT01864603). We discuss how missing data can be viewed as a causal inference problem and the use of machine learning to improve control of differential missingness. We conclude with the use of machine learning to improve precision and statistical power when assessing comparative effectiveness in randomized trials.

Laura B. Balzer, PhD

Meet the Speaker

Dr. Laura Balzer is an Assistant Professor of Biostatistics at the University of Massachusetts-Amherst. She is the Director of the UMass Causality Lab, and her areas of expertise include Causal inference and Machine Learning. These disciplines are integral to developing, evaluating, and implementing data-driven solutions in Public Health and Medicine. Dr. Balzer is the Primary Statistician for three cluster randomized trials in East Africa: the SEARCH study to prevent HIV and improve community health, the SATURN study to improve care outcomes among HIV+ youth, and the SPIRIT study to prevent TB. Her work is supported by the National Institutes of Health (NIH) and has been recognized with the ASA’s Causality in Statistics Education Award.

This activity has been approved for AMA PRA Category 1 Credit

Stay Connected

Join the ACCEL newsletter and keep up to date with the latest news, events and funding opportunities!

[yikes-mailchimp form="1"]