You are here

Constructing survival analysis models with independent variables that change over time: Application using data from the Survey of Labour and Income Dynamics

TitleConstructing survival analysis models with independent variables that change over time: Application using data from the Survey of Labour and Income Dynamics
Year of Publication2004
AuthorsGonthier, D.
JournalThe Research Data Centres Information and Technical Bulletin
Volume1
Pages6 - 12
Keywordsanalytical techniques, statistical methods
Abstract

This article provides a practical example of the development of a survival analysis model. It begins with an overview of the software tool that was used, SAS. The next section examines the construction of a longitudinal file and the challenges that may present. Of particular interest are explanatory variables that do not have a constant value over time. An example of a practical application is provided to illustrate the survival approach. The example consists of an analysis based on data from the Survey of Labour and Income Dynamics (SLID), specifically data from panel 1 between January 1993 and December 1998. Survey information in vector form is used to develop a Cox semi-parametric model. Comments are provided on a sample computer program. The way in which the program handles the main variables is also discussed. The last section contains a brief description of the results of a relatively simple model.

URLhttp://www.statcan.gc.ca/pub/12-002-x/12-002-x2004001-eng.pdf