Click on this small photo to view larger one.

Survival Analysis: Finding the Correct Distribution for Survival Data Faculty Advisor: Robin Lock Format: Poster

This project examines survival analysis with a focus on estimation techniques for fitting lifetime distributions to survival data.  Some examples of survival data may include life spans after medical treatment, times until mechanical components fail, or length of marriages.  There are several methods and ways to estimate parameters.  The parametric models studied include: (1) exponential, (2) lognormal, (3) gamma, and (4) Weibull.  The (5) Kaplan-Meier estimator is an alternative nonparametric technique that is used to fit survival data.  When trying to fit a lifetime distribution, censoring may become an issue.  Complete data sets include all information for each subject, and censoring occurs when subjects leave the study or outlast the length of the study.  For each model, these techniques are illustrated and compared with applications to simulated and real data.