As of Jan 1, 2023, I am Professor in the Division of Biostatistics at the University of Minnesota and Director of Biostatistics at the University of Minnesota Masonic Comprehensive Cancer Center. From 2010-2022, I was the Director of the Division of Biostatistics in the Department of Public Health Sciences at the University of Miami, Miller School of Medicine. From 2016 (June) - 2019 (December), I was the Interim Chair of the Department of Public Health Sciences. More on my time in that role can be found here.
From 1998-2010 I was in the Department of Epidemiology and Biostatistics at Case Western Reserve University School of Medicine where I rose to Full Professor. For the last 5 of those years, I was Director of the Division of Biostatistics. From 1994-1998 I was on faculty in the Department of Biostatistics at the Cleveland Clinic Foundation.
I graduated from the University of Toronto in 1994 with my Ph.D. in Biostatistics under the guidance of Rob Tibshirani. In 1991, I received my M.S. degree in Biostatistics from the University of Minnesota and in 1989, I received my BSc. from the University of Ottawa with a double major in Biology and Biochemistry. I am a Fellow of the American Statistical Association (2011), Elected Member of the International Statistical Institute (2016), a Fellow of the Institute of Mathematical Statistics (2024), and an Honorary Member of the Society of Statistics, Computer and Applications (2024).
Research Interests
My work is mostly motivated by real problems I encounter from medicine and biology. Much of my time has been in the area of cancer genomics where I co-developed Bayesian ANOVA for microarrays (BAMarray) and bump hunting for discovery of novel subgroups of colon cancer. I was also part of the research team that conducted the original work that became Cologuard - the at-home, early detection of colon cancer screening tool based on detection of aberrant DNA methylation in stool. I am also interested in modeling pharmacogenomic data looking for strategies for identifying candidates for drug repurposing drug synergies. I am now working in developing new statistical methods for health disparity estimation. I recently co-developed PRISM for multilevel tree-based modeling of individual level and social determinants of health and disparity driver analysis (DDA). Most recently, I've also gotten interested in developing new models to estimate contextual vulnerability for better predicting and risk stratifying patients for opioid relapse. And I too am researching various aspects of COVID-19 modeling including correcting prevalence estimation for biased testing sampling.
Other areas of statistical research include a more continuous form of spike and slab regression for high dimesnional L2 shrinkage with hard thresholding, model selection for complex data including fence methods for linear and generalized linear mixed models and non-parametric small area estimation, and the E-MS algorithm for model selection with missing or incomplete data. I have also done work on mixed model prediction (MMP) which includes the development of the observed best predictor (OBP) and best predictive estimator (BPE) for misspecified mixed models, and classified mixed model prediction (CMMP) for more accurate subject level prediction. These have all been recognized as important contributions to statistics.