A latent class growth model for migrants’ remittances: an application to the German Socio-Economic Panel

Aufsätze referiert extern - Web of Science

Silvia Bacci, Francesco Bartolucci, Giulia Bettin, Claudia Pigini

In: Journal of the Royal Statistical Society: Series A (Statistics in Society) 182 (2019), 4, 1607-1632

Abstract

We propose a latent class mixture growth model with concomitant variables to study the time profiles of international remittances sent by first-generation migrants in Germany from 1996 to 2012. The latent class approach enables us to identify homogeneous subgroups of migrants associated with different trajectories for their remitting behaviour, which can be interpreted in the light of the theoretical economic background. In addition, the inclusion of concomitant covariates allows us to uncover whether the assignment of migrants to a specific subgroup can be ascribed to their observable characteristics (e.g. their intention to return home), as conjectured by the theoretical models. The model proposed is easily estimated through an expectation–maximization algorithm. Results show that migrants can be clustered in three groups, two of which reflect the evolution of remittances predicted by economic theory.



Keywords: Concomitant variables approach; Latent class model; Latent trajectory model; Longitudinal data; Remittances
DOI:
https://doi.org/10.1111/rssa.12475