Flexible Transition Timing in Discrete-Time Multistate Life Tables Using Markov Chains with Rewards

Daniel C. Schneider , Max Planck Institute for Demographic Research
Mikko Myrskylä, Max Planck Institute for Demographic Research and London School of Economics and Political Science (LSE)
Alyson A. Van Raalte, Max Planck Institute for Demographic Research (MPIDR)
Hal Caswell, University of Amsterdam
Silke van Daalen, University of Amsterdam

A standard tool in demographic research are continuous-time multistate life tables (MSLTs). More recently, similar methods have been advanced that are cast in discrete time. While these models use a discrete time grid, it is often useful to calculate derived magnitudes, like state occupation times, under assumptions that posit that transitions take place at other times, such as mid-period. We propose to utilize Markov chains with rewards as an intuitive way of modelling the timing of transitions. This has the following advantages: It allows a flexible modelling of transition timing; carries a manageable theoretical overhead; and is easy to apply. We illustrate the usefulness of rewards-based MSLTs with SHARE data for the estimation of working life expectancy using different retirement transition timings. We also demonstrate that, for the single-state case, the rewards-based MSLTs match traditional life table methods exactly. We provide R and Stata code for the method proposed.

See paper

 Presented in Session 30. Life Tables and Applications