Dr. Hejazi's research interests concentrate around causal inference and statistical machine learning, focusing on the development of model-agnostic methods for statistical inference. This approach stresses developing techniques that are (1) robust, by leveraging data-adaptive and flexible estimation procedures; (2) statistically efficient (i.e., minimal variance), by using cutting-edge semi-parametric efficiency theory; and (3) assumption-lean, by incorporating domain knowledge while also respecting its limitations. His problem-first, translational philosophy emphasizes tailoring inferential methods to concrete, well-specified scientific questions, ensuring that statistical methods contribute directly to expanding scientific knowledge. Dr. Hejazi is often motivated by topics from non- and semi-parametric (model-agnostic) inference and efficiency theory; high-dimensional inference; (targeted) minimum loss-based estimation; uses of and corrections for biased sampling procedures (e.g., outcome-dependent two-phase designs); and adaptive experimental designs (e.g., to ascertain sequentially adaptive optimal treatment schemes). Most often, he studies these topics through the lens of causal inference, which formalizes scientific inquiries via parameters that can be endowed with causal interpretations (e.g., heterogeneous treatment effects, dose-response curves, mediational direct/indirect effects). Dr. Hejazi is also deeply interested in high-performance statistical computing and is a passionate advocate of open-source software for the promotion of transparency, reproducibility, and "data analytic hygiene" in the practice of applied statistics and statistical data science.
Dr. Hejazi's substantive scientific interests span diverse areas of the biomedical and public health sciences, and he has participated in scientific collaborations across a broad array of disciplines, including molecular and environmental toxicology, large-scale interventional nutritional epidemiology, comparative (health) effectiveness research using electronic health/medical records, and computational/high-dimensional biology. Most recently, he has been captivated by the rich statistical issues and pressing public health challenges common in clinical trials and/or observational studies evaluating the efficacy of preventive vaccines or curatives/therapeutics for high-burden infectious diseases (HIV/AIDS, COVID-19), in infectious disease epidemiology, and in computational immunology.
Some core elements of Dr. Hejazi's philosophy of statistics have already been more eloquently expressed by others:
"Far better an approximate answer to the right question, which is often vague, than the exact answer to the wrong question, which can always be made precise." --John Tukey
"Everyone is sure of this [that errors are normally distributed]...since the experimentalists believe that it is a mathematical theorem, and the mathematicians that it is an experimentally determined fact." --Henri Poincare
BA, 2015, Molecular & Cell Biology, Psychology, Public Health
University of California, Berkeley, Berkeley, CA, USA
MA, 2017, Biostatistics
University of California, Berkeley, Berkeley, CA, USA
PhD, 2021, Biostatistics
University of California, Berkeley, Berkeley, CA, USA
NSF Postdoctoral Research Fellowship, 2022, Biostatistics, Causal Inference, Machine Learning
Weill Cornell Medicine, New York, NY, USA
The Wellness Scholarship in Honor of Chin Long Chiang2018-2018
UC Berkeley School of Public Health
The Eki & Nobuta Akahoshi and Seiko Baba Brodbeck Endowed Fund Scholarship2019-2019
UC Berkeley School of Public Health
Tom Ten Have Memorial Award (for "exceptionally creative or skillful research in causal inference")2019-2019
American Causal Inference Conference
The Wallace Lowe Fellowship2020-2020
UC Berkeley School of Public Health
NSF Mathematical Sciences Postdoctoral Research Fellowship2021-2022
National Science Foundation