Individual heterogeneity is now a fundamental component of statistical and econometric models. Moving away from representative agent models and the estimation of average treatment effects, heterogeneous agents and heterogeneous treatment responses are at the core of modern analyses. However, such heterogeneity is often unobserved, latent, or poorly measured. To overcome this measurement issue, we intend to leverage genetic information. Genetic variants provide a biologically based, potentially causal, and observable measure of heterogeneity that can be easily incorporated into statistical and econometric models. The project will build two general methods of improving our modelling of individual-level heterogeneity by using genetic variants. The first one focuses on structural models of health and human capital formation; the second is an econometric estimation of treatment effects. The first method will combine economic theory and structural models with genetic variants. Starting from a workhorse model of health and human capital formation, the project will consider which genetic variants—genotypes—could be linked to the deep structural parameters of the model. This will provide a biological foundation to the heterogeneity in the individual’s choices. The second method aims at merging two streams of literature: the debate on nature and nurture and the potential outcomes framework. This approach explicitly models the way genetic predisposition can shape selection into treatment (connecting it to what behavioural geneticists call gene-environment correlation; rGE), and heterogeneity of treatment effects (connecting it to gene-environment interactions; GxE).
This project is hosted at University of Bologna.
This project has received funding from the European Union’s HORIZON-MSCA-2021-DN-01 programme under grant agreement number 101073237
essgn.org 