Survival models and health sequences: discussion
- 103 Downloads
“I can’t remember things before they happen”
“It’s a poor sort of memory that only works backwards”
— Exchange from “Alice Through the Looking Glass”
I thank the editors for the opportunity to comment on this interesting and challenging paper. The insight that death is a part of life, its ultimate destination, has mathematical as well as philosophical, content. The paper also makes the important general point that the full specification of the joint distributions of a stochastic process can be decomposed (factorized) in many different ways and that the familiar approach based on its forward evolution through time does not always provide the simplest description of its structure. I found the cited paper by Kurland et al. (2009) very helpful in this regard. I would like to make a pitch for what these authors call the “partly conditional approach” which first models the time to death and subsequently the conditional distribution of the health status process among survivors, perhaps to...
- Efron B (1967) The two-sample problem with censored data. In: leCam L, Neyman J (eds) Proc. 5’th Berkeley Symp Math Statist Prob, vol 4. University of Califormia Press, Berkeley, pp 831–853Google Scholar
- Kong S, Nan B, Kalbfleisch JD, Saran R, Hirth R (2017) Conditional modeling of longitudinal data with terminal event. J Am Stat Assoc 71:340–344Google Scholar