An Actuarial Survey of Statistical Models for Decrement and Transition Data - I: Multiple State, Poisson and Binomial Models
- 1 April 1996
- journal article
- Published by Cambridge University Press (CUP) in British Actuarial Journal
- Vol. 2 (1) , 129-155
Abstract
This paper surveys some statistical models of survival data. A basic model of a random lifetime is defined, and censoring is introduced. Methods based on observations of small segments of lifetimes are compared. Markov and semi-Markov (multiple state) models are recommended as well-understood and flexible models well suited to actuarial data. A Poisson model is discussed as an approximation to a two state model, while traditional Binomial-type models are shown to be more restricted and less tractable than multiple state models.Keywords
This publication has 43 references indexed in Scilit:
- An Actuarial Model for AIDSJournal of the Staple Inn Actuarial Society, 1990
- The analysis of event history data: A review of progress and outstanding problemsStatistics in Medicine, 1988
- A short note on the construction of life tables and multiple decrement tablesJournal of the Institute of Actuaries, 1987
- Dynamic modelling and causalityScandinavian Actuarial Journal, 1987
- A flaw in actuarial exposed-to-risk theoryScandinavian Actuarial Journal, 1984
- Understanding Cox's Regression Model: A Martingale ApproachJournal of the American Statistical Association, 1984
- Identifiability for dependent multiple decrement/competing risk modelsScandinavian Actuarial Journal, 1983
- Multistate Life Table Methodology: A Probabilist Critique1982
- Fitting Cox's Regression Model to Survival Data using GLIMJournal of the Royal Statistical Society Series C: Applied Statistics, 1980
- A nonidentifiability aspect of the problem of competing risks.Proceedings of the National Academy of Sciences, 1975