Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.
Author ORCID Identifier
Campus-Only Access for Five (5) Years
Doctor of Philosophy (PhD)
Year Degree Awarded
Month Degree Awarded
Biostatistics | Statistical Methodology | Statistical Models | Survival Analysis
Nested case-control study design, which is cost-eﬀective, has been widely used in large prospective epidemiological studies. In these studies, biomarkers exhibiting time-varying associations with disease outcome occur frequently. To analyze data from nested case-control study design with temporal biomarker eﬀects, we proposed a martingale-based estimating procedure building on a generalized Cox model, an estimating equation approach to a censored quantile regression model, and a kernel weighted pseudo-likelihood approach on the classic Cox model. The inverse probability weighting technique is implemented to overcome the complex correlation structure among nested case-control sub-cohort. A perturbation based resampling procedure is proposed for variance estimation, and furthermore, a computationally eﬃcient resampling procedure is developed. We study the asymptotic properties of the resulting estimators, including the uniform consistency and weak convergence. Extensions with additional matching and second-stage inferences are developed accordingly. Simulation studies demonstrated nice ﬁnite sample performance of the proposed methods and procedures. Real data analysis on Nurses’ Health Study and Wilms Tumor Study are presented for illustration.
Zhang, Yiding, "REGRESSION ANALYSIS OF TEMPORAL BIOMARKER EFFECTS UNDER NESTED CASE-CONTROL STUDIES" (2021). Doctoral Dissertations. 2235.