Independent Research Fund Denmark (DFF)
Causal Inference with Multiple IVs: Using Mendelian Randomization to Estimate the Causal Effects of Cannabis and Alcohol on Labor Productivity and of Creativity and Risk Taking on Entrepreneurship
This project will develop new instrumental variable (IV) methods that will be relevant, not only for empirical researchers in economics and the social sciences, but in many other disciplines, including epidemiology and genetics. The relevance of the new proposed methods will be illustrated by two empirical applications using the unique IPSYCH genetic dataset. Adapting the newly developed IV methods to the idea of Mendelian randomization, I will use genetic variants as instruments to i) estimate and compare the causal effects of cannabis and alcohol abuse on educational attainment, labor market outcomes, and health outcomes and ii) estimate the causal effects of creativity and risk propensity on the probability of engaging in entrepreneurial activities.