Model Survival Data with high and ultrahigh dimensional predictors: feature screening and statistical inference - Lecture 2

Abstract

Continuing on with Lecture 1,  this short course introduces various cutting-edge methods that handle survival outcome data with ultrahigh dimensional predictors, that is when the dimension of predictors is much higher than the sample size. We will also discuss several new methods for quantifying the uncertainty of estimates in a high-dimensional survival setting, a very active area in machine learning.  

The specific topics include.

Feature screening with ultra-high dimensional predictors (p>>n):

1. Principled sure independent screening (PSIS),
2. Conditional screening,
3. IPOD,
4. Forward selection, etc;

Inference for survival models with high dimensional predictors (p>n).

Brief Biography

Yi Li is a Professor of Biostatistics and Professor of Public Health. His current research interests are data science, survival analysis, longitudinal and correlated data analysis, measurement error problems, spatial models, and clinical trial designs. His methodologic research is funded by various NIH statistical grants starting from year 2003. Yi Li is actively involved in collaborative research in modern biomedical studies with researchers from the University of Michigan and Harvard University. The applications have included chronic kidney disease surveillance, organ transplantation, cancer preventive studies and cancer genomics. He has made fundamental contributions to predictive modeling and risk assessment, publishing more than 200 papers in leading statistical journals, such as JASA, JRSSB, Biometrika, and Biometrics, and medical journals, such as JAMA, PNAS, and JCO.