Ming Wang, PhD
Associate Professor, Department of Population and Quantitative Health Sciences
School of Medicine, Case Western Reserve University
Abstract: Complex and big data are increasingly available nowadays and encompass many scientific fields including cancer. Motivated by clinical needs and complexities of data features that cannot be handled by conventional methods, advanced statistical models are in demand for development to achieve valid and robust inference. In this talk, I will showcase several recent method developments with potential applications to cancer and other diseases. From these works, we formulated novel methods for longitudinal and survival data analysis by addressing a variety of data issues including competing risk, missing data, subject heterogeneity, and others. Rigorous theoretical justifications were fully investigated. In addition, we used large data resources from observation studies (e.g., the Cardiovascular Health Study) and public data registries (e.g., the National Cancer Database; the Surveillance, Epidemiology, and End Results Program) for method illustration and clinical exploration.