Course Outline:
Bayesian methods have become increasingly popular, especially in macroeconomics. The large dimensionality of macro-econometric models and the complexity of modern DSGE models often require the use of prior information and computational algorithms to conduct econometric inference. This course will give an introduction to Bayesian estimation both from a technical and practical point of view. The curriculum will cover basic notions of Bayesian inference and posterior simulators, with applications to regression and state space models. Empirical applications and more advanced topics will be treated in reading groups.
Although the focus of the course is on macro-oriented models, micro-oriented student presentations are encouraged.
This course is tailored towards advanced masters and graduate students in Economics or other related disciplines.
Learning Outcomes:
Students should
- Become familiar to key concepts of Bayesian Inference and its differences from Frequentist inference
- Be able to construct meaningful priors and be aware of the effects of prior information on inference
- Understand main posterior sampling techniques, and how to summarize posterior information
- Be knowledgeable of more advanced topics in econometric research on Bayesian methods and their applications to Economics
Details:
Prerequisites: Students should have basic knowledge of probability, regression, time series (ARMA modeling etc) and scientific programming. Familiarity with modern dynamic macroeconomic models is desirable.
Contact schedule:
- 2 hours of lectures per week (4 weeks)
- Lecture 1: Introduction to Bayesian Inference
- Lectures 2-4 : Prior elicitation and Posterior sampling algorithms with focus on Regression based
- models, State Space (SS) models
- Reading group- Presentations
- The group will meet once weekly for two hours and discuss assigned papers.
- Meeting Times: TBA
Evaluation:
The course evaluation will be based on paper/book presentations and reports.
Restriction to participation: 20
Registration: in the first lecture |
