Skip to main content

A latent Markov model with two parallel processes for modeling inter-generational exchanges

Headshot of Irini Moustaki

Irini Moustaki


In this paper, we model a type of dyadic data that provides information on inter-generational help exchanges in the UK. We use longitudinal data from five waves from 2011 to 2021 of the UK Household Longitudinal Survey (UKHLS) to study and explain associations between support exchanges between adult children and parents who do not live together. The questions analyzed here are from the family network module of the UKHLS. The survey respondents report exchanges of help from a child's perspective about a dyad. The dyad is defined as help from children to parents and received from parents. The data resemble the structure of dyadic data; they are collected across time and are also multivariate because the level of support is measured by multiple indicators (a set of eight binary indicators of different kinds of help that, in most cases, require the helper's time). 

We propose a Hidden Markov cross-lagged joint model of bidirectional exchanges with support given and support received treated as multivariate responses and covariances between responses measuring the extent of reciprocation between generations.  Moreover, joint modeling of longitudinal data allows for the possibility that reciprocation may occur contemporaneously or may be postponed until the donor needs help or the recipient is in a position to reciprocate.

Irini Moustaki is a Professor in the Department of Statistics at the London School of Economics & Political Science.