A Partly Parametric Additive Risk Model
Abstract
Aalen's additive risk model allows the influence of covariates on a hazard function to vary over time, and to do so in a different fashion for each covariate. Although allowing greater flexibility than a Cox model, which has a more parsimonious temporal structure, the number of covariates that can be handled by Aalen's model is quite limited. One way around this difficulty is to impose some a priori structure on the form of the model, thereby reducing the number of functions to be estimated. In this paper we introduce a partly parametric version of Aalen's model in which only a small number of the covariates are selected to have their influence vary nonparametrically over time, and the influence of the remaining covariates is restricted to be constant in time. Efficient procedures for fitting this new model are developed and studied. The approach is applied to data from the British Medical Research Council's myelomatosis trials.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1993
- Accession Number
- ADA260395
Entities
People
- Ian W. Mckeague
- Peter D. Sasieni
Organizations
- Florida State University