031. Discrete Time to Event Data

In this video we formally introduce the idea with discrete time survival analysis, introducing both intuitive/historical estimators for the quantities of interest, as well as a discussion of how an extension of transition models can be used to characterize the likelihood in such a way so as to be able to estimate the quantities (even when we have covariates present). The formalization of these ideas are left for a later date, when we introduce the proportional odds models directly.

Video Timeline
00:00 - Introduction
01:05 - Brief Example
04:58 - The Hazard Function
09:44 - The Survivor Function
12:22 - Measures of Central Tendency
16:33 - Likelihood Based Analysis