Type
Text
Type
Dissertation
Advisor
Reich, Nancy C. | Nanct Medell R. Mendell | Wei Zhu | Derek Gordon.
Date
2010-08-01
Keywords
Statistics -- Biology, Biostatistics | exponential survival analysis, Mixture Survival Analysis, Quandt Ramsey
Department
Department of Applied Mathematics and Statistics
Language
en_US
Source
This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.
Identifier
http://hdl.handle.net/11401/70907
Publisher
The Graduate School, Stony Brook University: Stony Brook, NY.
Format
application/pdf
Abstract
The mixture of two regression regimes has been extensively studied in economics. A switching regression is often used to model a system that changes depending on some variables. The test of a mixture of regimes in hazard modeling would be seen to have fundamental importance in biostatistical research but has not been studied. A two-regime parametric mixture is proposed to model the effect of a single covariate on the event time. Typically, the Cox proportional hazards model is applied to estimate a single regime survival regression function. The mixture of two regimes model contains five parameters to be estimated; namely, two parameters to describe each regime, and one to describe the mixing proportion. A software program developed for this research finds the maximum likelihood estimates of the parameters and the likelihood ratio test of the null hypothesis of a single regime against the alternative of a mixture of two regimes. A simulation study finds an approximation to the null distribution of the test and its approximate power.
Recommended Citation
Chen, Paichuan, "Extending the Quandt-Ramsey Modeling to Survival Analysis" (2010). Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions). 115.
https://commons.library.stonybrook.edu/stony-brook-theses-and-dissertations-collection/115