Advanced Econometrics I: Non-Linear Cross-Sectional Models
Advanced econometrics has been reorganized into three 5-week courses (Advanced Econometrics I, II and III) that are offered according to student interest and can be taken individually or as a sequence. Together, the 15-week Advanced Econometric series provides students with a toolkit to understand and conduct applied research in the social sciences.
The required Empirical Analysis III course covered linear models and the identification of causal effects. Linear models are fundamental tools and often sufficient for basic applications, but many research problems require different approaches. Outcomes rarely have a nice continuous distributions, many responses cluster on a certain value (say zero, e.g. doctor visits), or are top-coded (e.g. income reported in tax returns). Others are binary by nature (decision to work or not to work) or are ordered, but not a cardinal scale (e.g. health reported excellent/very good/fair).
Advanced Econometrics I covers the most standard nonlinear statistical models and the basic theory behind it. These include: Maximum Likelihood Estimation; Properties of MLE and other nonlinear methods; Consistency; Asymptotic normality; Efficiency; and Numerical Optimization and Bootstrapping.