Focussed model selection and model averaging for the Cox regression model

Add to your list(s) Download to your calendar using vCal

• Gerda Claeskens, Katholieke Universiteit Leuven
• Tuesday 03 June 2008, 14:30-15:30
• Large Seminar Room, 1st Floor, Institute of Public Health, University Forvie Site, Robinson Way, Cambridge.

If you have a question about this talk, please contact Nikolaos Demiris.

Survival data are recorded consisting of life times, censoring indicators, and a whole set of covariates of which not all of them might be worth including in an analysis. A popular model for such data is the Cox model of proportional hazards. Inside this semiparametric model we study the variable selection problem. Models are constructed for various reasons: to estimate the relative risk, a survival probability, a cumulative hazard,... The focussed information criterion provides a way to select the best model depending on this quantity of interest. We also discuss the important issue of inference after model selection, dealt with via model averaging estimators. (This is joint work with Nils Lid Hjort.)

This talk is part of the MRC Biostatistics Unit Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Immunology
• Cambridge Immunology
• Large Seminar Room, 1st Floor, Institute of Public Health, University Forvie Site, Robinson Way, Cambridge
• Life Sciences
• Life Sciences
• MRC Biostatistics Unit Seminars

Note that ex-directory lists are not shown.