Computational Economics - Advanced Level
Diese Veranstaltung richtet sich an Studierende der Master Studiengänge „Economics“ und „Wirtschaftsmathematik“. Sie ersetzt die frühere Veranstaltung „Numerische Gleichgewichtsmodelle“. Die Veranstaltung kann generell in den Schwerpunktfächern „Finanzwissenschaft“ und „Forschungsmethoden“ angerechnet werden. Studierende in den Studiengängen „Wirtschaftsmathematik“ sowie „Economics“ nach der alten Prüfungsordnung können sie zusätzlich auch im Schwerpunktfach „Ökonometrie“ anrechnen lassen.
Syllabus: Computational Economics - Advanced Level
Aim of the course:
This course will mostly be concerned with analyzing public policy (like taxation, social security, etc.). We will learn how such policies redistribute between different generations and also within generations, how they may improve risk sharing when markets ar incomplete and how they can trigger distortions and therefore hurt the aggregate economy. To accomplish this task, this course provides students with state of the art techniques for quantitative maroeconomic research in this very field and familiarizes them with the relevant literature. Ideally at the end of the course, they will be able to develop their own ideas and conduct their own policy analysis using the techniques acquired.
The course will consist of a series of lectures and exercise classes. The teaching sequence is divided into three parts:
- In the first part, students will review programming in FORTRAN and applying numerical methods. Note that it is expected here that students have already some programming skills.
- The second part will be concerned with solution techniques to dynamic programming problems and will introduce the basics of the model we are concerned with most of the time: the overlapping generations model (OLG) with uninsurable income risk.
- In the last part focusses on reliable policy analysis in the stochastic OLG model and discusses recent literature that applies this model to analyze public policy reforms.
Students that attend this course should know about basic macroeconomics. Knowledge of heterogeneous agent models will be helpful. As already said, some programming skills are required for this course. The BA course "Computational Economics" would be perfect, but also some background in Matlab programming is fine. In addition, the should bring along a strong willingness to specialize in programming (which implies that they will program a lot themselves).
Grades will be based on student's participation in the course as well as on some programming assignments.
There will be ample course material on how to program in FORTRAN, which compilers to use, numerical techniques, etc. In addition to a couple of chapters on these topics, there will be FORTRAN codes available for everything we do. There is no baook, but lecture notes will be provided in pdf format.