My main research area has been time-to-event data analysis and clinical trials. In time-to-event data analysis, I have worked on frailty modeling, efficiency of survival probability estimates from the proportional hazards model, weighted log-rank test, competing risks, quantile residual lifetime/life lost, and likelihood theory such as empirical likelihood and hierarchical likelihood.
In clinical trials, I have been involved in several influential phase III breast cancer clinical trials, including work on developing a prediction model based on microarray data collected from different platforms, through the National Surgical Adjuvant Breast and Bowel Project (NSABP) as a co-investigator. My current active collaboration is focused on pain medicine, developing prediction models and facilitating clinical trials. I also serve as a senior faculty member in the CTSI Biostatistics, Epidemiology, and Research Design (BERD) Core at University of Pittsburgh.
My recent method and collaborative research efforts have evolved toward the areas of machine learning and causal inference in survival analysis. Specifically, a deep learning algorithm has been developed for the quantile regression for censored time-to-event data, and the random forest algorithm has been applied to predict an optimal combination of therapies for patients in pain.
I graduated 6 MS and 11 PhD students, and am currently advising 1 PhD student. I have been teaching Statistical Theory, Survival Analysis, and Mixed Models, all at both MS and PhD levels. I currently serve on the editorial board for Lifetime Data Analysis, the only journal dedicated to time-to-event data analysis. I am a Fellow of the American Statistical Association (ASA), an elected member of the International Statistical Institute (ISI), and an elected member of Omicron of the Delta Omega Society (Honorary Public Health Society).