8 Fitting Cox Models in Practice
A Cox model in a real study includes several covariates of different kinds, has to deal with tied event times, may need interactions, and is used to predict survival for particular patients. This section works through these practical matters with the veteran lung cancer trial, in which \(137\) men were randomised to a standard or a test chemotherapy, with tumour cell type and the Karnofsky performance score (from \(10\), moribund, to \(100\), normal) recorded for each.
8.1 Categorical covariates
8.2 Tied event times
8.3 Interactions and stratification
8.4 Model building
8.5 Adjusted survival curves
8.6 Model fitting in software
8.7 Practice Problems
8.2 Tied event times
8.3 Interactions and stratification
8.4 Model building
8.5 Adjusted survival curves
8.6 Model fitting in software
8.7 Practice Problems
Questions on this section
Stuck on something here? Ask below and it stays attached to this topic.