Yun Li, PhD

Dr. Li conducts methodological research in causal inference, unmeasured confounding, missing data, mediation, Bayesian analyses and survey methods. She has led the effort in database management, data integrity, rigorous and robust analytical methods in many multicenter studies and NIH-funded program projects in biomedical science, in areas including including breast cancer, prostate cancer, head-and-neck cancer, kidney disease, cardiovascular diseases and liver disease in the past two decades.

Needs that arose in the course of her collaborative work have inspired her to innovate in statistical methods research, and she has led the development of statistical methods to tackle the statistical issues in these clinical studies. Her methodological work has involved four key areas: 1) intermediate outcomes; 2) time-dependent treatments; 3) unmeasured confounders; and 4) missing data. Of note, her early methods work was among the first to propose a causal inference framework for evaluating surrogate endpoints, such as intermediate biomarkers, surrogate markers or mediators.

Prior to her work at Penn, Dr. Li was a Research Associate Professor at the University of Michigan (UM). Dr. Li earned her PhD in Biostatistics from UM, following a period of highly effective collaborative work at Duke University Clinical Research Institute. She joined the UM Biostatistics faculty as a Research Assistant Professor after graduation.

Content Area Specialties

Infectious disease and pediatric medicine, breast cancer, kidney disease, cardiovascular disease.

Methodology Specialties

Causal inference, unmeasured confounding, missing data, mediation, Bayesian analyses and survey methods

Recent Publications

  1. Lehmann D, Li Y, Saran R, Li Y: Strengthening instrumental variables through weighting. Statistics in Biosciences 9(2): 320-338, 2017 Notes: doi:10.1007/s12561-016-9149-9. Senior authorship; my doctoral student's dissertation.
  2. Elliot MR, Conlon ASC, Li Y, Kaciroti N, Taylor JMG: Surrogacy paradox measures in meta-anlaytic settings Biostatistics 16(2): 400-12, Apr 2015.
  3. Li Y, Lee Y, Wolfe RA, Morgenstern H, Zhang J, Port F and Robinson BM: On a preference-based instrumental variable approach in reducing unmeasured confounding-by-indication. Statistics in Medicine 34(7): 1150-68, Mar 2015.
  4. Li Y, Schaubel DE and He K: Matching methods for obtaining survival functions to estimate the effect of a time-dependent treatment. Statistics in Biosciences 6(1): 105-126, May 2014.
  5. Elliott MR, Conlon A, Li Y: Discussion of "Surrogate measures and consistent surrogates " Biometrics 69(3): 565-569, Sep 2013.
  6. Li Y, Taylor JMG, Elliott MR and Sargent D: Causal assessment of surrogacy in a meta analysis of colorectal cancer trials. Biostatistics 12(3): 478-492, Jul 2011.
  7. Li Y, Taylor JMG and Little RJA: A shrinkage approach for estimating a treatment effect using intermediate biomarker data in clinical trials Biometrics 67(4): 1434-1441, Dec. 2011.
  8. Li Y and Taylor JMG: Predicting treatment effects using biomarker data in a meta-analysis of clinical trials. Statistics in Medicine 29(18): 1875-1889, July 2010.
  9. Elliott MR, Raghunathan N, Li Y: Bayesian inference for causal mediation effects using principal stratification with dichotomous mediators and outcomes. Biostatistics 11(2): 353-372, Apr 2010.
  10. Li Y, Taylor JMG, Elliott MR: A Bayesian approach to surrogacy assessment using principal stratification in clinical trials. Biometrics 66(2): 523-531, Jun 2010.