Empirical Analysis III: Econometrics
Often in policy analysis, the crucial analytic question concerns the causal effect of a policy, program, or procedure. This course provides an in-depth consideration of the problem of estimating causal effects and the three leading approaches to doing so in applied work: random assignment, fixed effects, and instrumental variables. To do so, the course develops linear asymptotic theory, which is applied to analyzing causal models and then to analyzing data that is not independent and identically distributed (non-i.i.d.). Non-i.i.d. models considered include heteroscedasticity, panel and clustered data, and time series. The course is taught through a combination of written problem sets, data analyses, and close reading of recent, exemplary papers.