Assistant Professor of Biostatistics
MD Anderson Cancer Center
email: jplong + the at symbol + mdanderson.org
Causal inference models make untestable assumptions regarding confounding variables. The impact of violations of these untestable assumptions on parameter inference can be severe. My methodological research in statistics focuses on building causal models to maximize causal prediction performance, the ability of the model to predict the effects of unseen interventions. This form of model evaluation is highly robust to violations in model assumptions and directly relevant many modern biological applications such as genetic knockout/knockout screens and design of rational combination therapies in cancer.
In addition to my methodological research, I collaborate will a large number of clinicians and biologists at MD Anderson on projects in cancer early detection, omics data analysis, and clinical trial design.
If you are a GSBS, UT Health, or Rice student interested in my research, please contact me.
See my google scholar profile for a list of publications. Code is available on github under longjp.
STAT 533 / GSBS 1283 - Foundations of Statistical Inference II: Spring 2020
Biostatistics 6082 - Survival Analysis: Spring 2019
Statistics 689 - Statistical Computing with R and Python: Spring 2018
Statistics 611 - Theory of Statistics II: Spring 2018
Astrostatistics: Astrostatistics course taught Fall 2016 at SAMSI.
Statistics 689 - Astrostatistics: Astrostatistics course taught Fall 2015 at TAMU.
November 10, 2021