Analysing plant closure effects using time-varying mixture-of-experts Markov chain clustering

Abstract

In this paper, we study data on discrete labor market transitions from Austria. In particular, we follow the careers of workers who experience a job displacement due to plant closure and observe -over a period of forty quarters -whether these workers manage to return to a steady career path. To analyse these discrete-valued panel data, we develop and apply a new method of Bayesian Markov chain clustering analysis based on inhomogeneous first order Markov transition processes with time-varying transition matrices. In addition, a mixture-of-experts approach allows us to model the prior probability to belong to a certain cluster in dependence of a set of covariates via a multinomial logit model. Our cluster analysis identifies five career patterns after plant closure and reveals that some workers cope quite easily with a job loss whereas others suffer large losses over extended periods of time.

Markov models for duration-dependent transitions: selecting the states using duration values or duration intervals?

“…A drawback of these approaches is that they do not allow for long-run analyses for the population as a whole since transitions between the age subgroups are not captured in the model making an analysis across age subgroups impossible. Instead, based on the Markov chains for each of the respective subgroups, the population as a whole can be modeled by a mixed Markov model (Langeheine and Van de Pol, 1994;Frühwirth-Schnatter et al, 2018).…”

Finite mixture model of hidden Markov regression with covariate dependence

“…While FMM can be used for partitioning the observations into similar groups, HMM can be applied to find homogeneous hidden states within those clusters. Finite mixture of HMM (FMM‐HMM) has been utilized in a wide range of applications including finance (Dias et al, 2009; Knab et al, 2003), transportation (Chamroukhi et al, 2011), clinical studies (Bartolucci et al, 2014; Marino & Alfò, 2020; Maruotti & Rocci, 2012) and labour market (Frühwirth‐Schnatter et al, 2018). On the contrary, Volant et al (2014) and Lin and Li (2017) have used mixtures as emission distributions in HMM.…”