Jennifer Starling
Ph.D. in Statistics, UT Austin (Spring 2020)
M.S. in Statistics, Texas A&M University
B.S. in Mathematics, Virginia Tech
Dissertation: "Bayesian methods for complex data structures, with applications to precision medicine in women’s healthcare"
Dissertation Advisor: James Scott
Research Interests: Bayesian methods, nonparametric regression, tree-based models, causal inference, public health, women's health, obstetrics
Website: https://jestarling.github.io
What does your research focus on? How is it applied?
My research interests are in developing flexible Bayesian models for structured data, tree-based methods, and causal inference, with applications to public health and clinical practice with a focus on women's healthcare and obstetrics.
This work includes developing models for nonparametric regression where the response is known to be smooth over a single covariate (often time). These methods are widely applicable across a variety of problems, and include complexities such as modeling smooth heterogeneous treatment effects, survival data, monotonicity, and accounting for uncertainty due to rounded responses.
My findings have added to clinical knowledge about the nature of stillbirth risk curves, the relative efficacy of a medical abortion protocol which lowers barriers to accessing care, and the relationship between pre-eclampsia and birth weight in low-resource obstetrics settings. These areas focus on providing clinicians and policy-makers information to make data-driven decisions on patient treatment and policy.
What research-related accomplishment are you most proud of and why? (paper, award, presentation, etc.)
I am especially proud of my work in estimating heterogeneous smooth treatment effects for comparing efficacy of two early medical abortion regimens. This research has contributed to lowering barriers to care access in the UK, in addition to giving important clinical insights about efficacy at later gestational ages. This work was recognized with the Thomas R. Ten Have Award from the Atlantic Causal Inference Conference.
What did you do before you entered the SDS Ph.D. program? What brought you to SDS at UT Austin?
Before entering the SDS PhD program, I spent several years working in financial services, and I completed my Masters in Statistics at Texas A&M University. Prior to that, I studied mathematics at Virginia Tech. I chose UT Austin because of the flexibility in choosing different research electives, and how many different research opportunities are available.
What is your favorite thing about SDS and/or UT-Austin? What are your post-graduation plans? Congratulations, by the way!
Thank you! One favorite thing is that SDS has been so supportive of me sharing my research -- I've been given many opportunities to travel, present my work, and meet friends and collaborators. After graduation, I have accepted a position as a Statistician with Mathematica Policy Research in Cambridge, MA. I am excited to join their Health focus group.
