Fitting Cox's Regression Model to Survival Data using GLIM

1980/01/01 by John Whitehead · 1 voice · 2 citations
Mathematics · #Statistical Distribution Estimation and Applications #Statistical Methods and Bayesian Inference #Statistical Methods and Inference

paper · doi:10.2307/2346901

crossref issued 1980/01/01 · crossref published 1980/01/01 · crossref published-print 1980/01/01 · openalex publication_date 1980/01/01 · crossref created 2006/06/18 · crossref deposited 2021/07/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/23 · crossref indexed 2026/07/28

Abstract

The proportional hazard regression model is reviewed, and its analysis using GLIM is described. Methods of estimating the underlying survivor functions are discussed. The Poisson model which allows the use of GLIM is introduced and interpreted. Two different treatments of tied observations are mentioned, and their properties are compared in the context of a specific example.

Cited by

• A Scalable Empirical Bayes Approach to Variable Selection
• Survival analysis as a classification problem

Discussions

• If you are willing to go away from the Cox model (by replacing the nonparametric hazard with a semiparametric component), you can use the life table approach. You bin the time axis and run a Poisson r [bsky, 0 points, 1 comments]