Main content

The Cumulative Effects of Time-Varying Covariates in Survival Analysis

Show full item record

Title: The Cumulative Effects of Time-Varying Covariates in Survival Analysis
Author: Lowe, Matthew
Department: Department of Mathematics and Statistics
Program: Mathematics and Statistics
Advisor: Horrocks, JulieDarlington, Gerarda
Abstract: Joint models for survival are modern statistical techniques for modelling the e ect of time-varying covariates measured with error on the hazard of survival. Previous research has demonstrated that joint models have less bias compared to traditional Cox models with time-varying covariates. This thesis compares three association structures for joint models: the current value speci cation, the cumulative speci cation, and the weighted cumulative speci cation. The three methods were compared on two real data sets. A simulation study compared the three speci cations when t to data generated according to the current value speci cation. Results from the simulation show that the current value and weighted cumulative speci cations provide similar estimates, with small bias and accurate standard deviation.
Date: 2020-11
Rights: Attribution 4.0 International
Terms of Use: All items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.

Files in this item

Files Size Format View
Lowe_Matthew_202012_MSc.pdf 630.7Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record

Attribution 4.0 International Except where otherwise noted, this item's license is described as Attribution 4.0 International