This is the home page of Pop509: Survival Analysis, being offered in the Spring of 2012 (Session I). For Pop 510: Multilevel Models, also offered in the Spring of 2012 (Session II), click here. For materials related to my research on multilevel models click here.
This half-course offered in the first half of the spring term focuses on the statistical analysis of time-to-event or survival data. We introduce the hazard and survival functions; censoring mechanisms, parametric and non-parametric estimation, and comparison of survival curves. We cover continuous and discrete-time regression models with emphasis on Cox's proportional hazards model and partial likelihood estimation. We discuss competing risk models, unobserved heterogeneity, and multivariate survival models including event history analysis. The course emphasizes basic concepts and techniques as well as applications in social science research using the statistical package Stata. Prerequisite: WWS509 or equivalent.
The slides I used in a workshop on survival analysis are available here.
For a more detailed description of the course, including a list of topics to be covered each week, see the syllabus.
Materials for week 1 include a handhout on Parametric Survival Models, a plot of the 2006 U.S. survival and hazard functions, and a Stata log fitting parametric models to recidivism data.
Weeks 2 and 3 are devoted to Non-parametric Estimation in Survival Models. Materials include a Stata log applying Kaplan-Meier and Mantel-Haenzsel, and a log fitting Cox's proportional hazards model to a two-group comparison. See also this application of Cox Regression to the recidivism data. We compare discrete and continuous time models fit to the same data.
Week 4 deals with Competing Risks, the analysis of survival time when there are multiple causes of failure. New materials include a note on cumulative incidence, including estimation of the cumulative incidence function (CIF) and Fine and Gray's competing risk model, and an expanded Stata log fitting competing risk models to the tenure of U.S. Supreme Court justices.
In week 5 we tackle Unobserved Heterogeneity, discussing univariate frailty models and the identification problem, including a very useful for mulas to conver back and forth between subject-specific and population-average hazards.
Week 6 is devoted to Multivariate Survival, where we review various approaches to the analysis of multiple-spell survival data, focusing on shared-frailty models. Don't miss the Stata handout fitting a shared frailty model to child survival data from Guatemala. I have added a discussion of model interpretation via post-estimation.