Andrew Titman - Testing the Markov assumption in general multi-state models
|Dates:||5 February 2020|
|Times:||14:00 - 15:00|
|What is it:||Seminar|
|Organiser:||Department of Mathematics|
|Who is it for:||University staff, External researchers, Current University students|
Andrew Titman (Lancaster University) joins us for the Statistics seminar series.
Recently there has been interest in the development of estimators of the transition probabilities for right-censored data that are robust to departures from the Markov assumption. The landmark Aalen-Johansen (LMAJ) 1 estimator is robust to non-Markov processes, but this robustness comes at the cost of a loss of efficiency compared to the standard Aalen-Johansen (AJ) estimator, making it important to identify when it is necessary to use LMAJ.
Three approaches to testing are considered; i) A simple method based on including time of entry into the state as a covariate in a Cox model for each transition intensity ii) Use of the stratified version of the Commenges-Andersen test 2 for a univariate frailty, and iii) A novel class of tests based on families of log-rank statistics, where patients are grouped by their state occupancy at landmark times.
All three approaches can be applied to both non-parametric and semi-parametric model. The performance of the test is investigated through simulation in a variety of settings and by application to datasets relating to liver cirrhosis and sleep behaviour.
1 Putter, H., Spitoni, C. (2018). Non-parametric estimation of transition probabilities in non-Markov multi-state models: the landmark Aalen-Johansen estimator. Statistical Methods in Medical Research. 27: 2081-2092.
2 Commenges, D. and Andersen, P.K. (1995). Score test of homogeneity for survival data. Lifetime Data Analysis, 1: 145-156.
Organisation: Lancaster University
Travel and Contact Information
Frank Adams 2
Alan Turing Building