Description:
This course deals with advanced inference in modern econometrics.
In the first part we review some elements of classical and Bayesian statistical theory, concentrating on special problems relevant to the study of econometrics: post-selection inference, multiple testing, and uniform asymptotics. Then we study extremum estimation problems (especially, GMM) with special attention to asymptotic theory and the weak instruments problem. The second part covers non- and semi-parametric topics including the bootstrap, density estimation, and non- and semi-parametric regression. The third part covers the concept of econometric identification, and possible frameworks to write down and interpret causal estimands (treatment effects). We also discuss a number of techniques for estimation of treatment effects (IV, Diff-and-Diff, RDD, Matching).
Literature:
see syllabus
Lecturers:
Gábor Uhrin
Time & venue:
Lectures/ tutorials: Tuesdays, 14:00-18:00; HU Berlin, Spandauer Str. 1, room 203
Exam:
Written exam (90 min)
More information can be found in the syllabus and on Moodle.
