Topics in Empirical Macroeconomics

Instructor:
Christiane Baumeister (Robert H. Lambert, Class of 1940, Helen B. Lambert, Mary E. Lambert and Michael P. Lambert Professor of Economics at the University of Notre Dame; Bundesbank Guest-Professor at Freie Universität Berlin)

Description:
Structural vector autoregressions are the workhorse models in empirical macroeconomics. The Bayesian approach to estimation and inference of (S)VAR models has gained popularity as models have become more complex. The goal of this course is to equip participants with the tools they need for state-of-the-art empirical research in macroeconomics and to develop practical skills to apply Bayesian methods to policy-relevant research questions. The first part of the course covers the basics of Bayesian econometrics including standard choices of prior distributions and numerical simulation methods. The second part of the course challenges the current practice of identification of VAR models by introducing a more general Bayesian framework that encompasses standard identification approaches as special cases. Drawing structural inference from VAR models requires making use of prior information. This course provides formal tools of Bayesian analysis that allow to incorporate prior beliefs about both the structural coefficients and the impacts of shocks in a flexible way and to characterize the contribution of prior information. The third part of the course extends the standard VAR model to allow macroeconomic dynamics to evolve over time and to incorporate a large cross section of variables. The course introduces state-space modeling as the common framework for both extensions. The methods introduced in the lectures will be illustrated with applications to the labor market, monetary policy, and oil price shocks in Matlab.

Literature:
see syllabus

Time & venue:
December 2-7, 2024; FU Berlin
For exact information on course timing and venue please see the course schedule.

Exam:
The final grade for the course is based on a take-home exam with programming exercises in Matlab. In addition, PhD students are required to give a 20-minute presentation of a research idea.

Registration:
Interested PhD students may register by email to buba@wiwiss.fu-berlin.de by October 31, 2024.

More information can be found in the course syllabus.