I was one of the first scholars to acknowledge the dynamic nature of migration behavior, in particular the importance of return migration. I have, together with collaborators, demonstrated that it is imperative to account for endogeneity of labor market (unemployment) and marriage dissolution when addressing their impact on return migration. We have also shown that it is necessary to correct for dynamic selection, due to return migration, for the analysis of economic assimilation of immigrants. Ignoring these dynamic selection issues of migration behavior severely biases the analyses.

An association between higher educational attainment and better health outcomes has been repeatedly reported in the literature. However, the extent to which education causes better health and later mortality is widely debated. Observed and unobserved factors, such as parental background and cognitive ability, confounds the analysis. One approach to account for this is using structural models in which the interdependence between education, health, and cognitive ability is explicitly modelled. I have, with collaborators, developed and estimated a structural model for (inherently non-linear) survival data (mortality and hospitalization). Ignoring the endogeneity of education would overestimate the impact of education on hospitalization and mortality.