Macroeconometrics 01: Introduction
About the course.
Our goal is to teach methods that will allow you to conduct an independent macroeconometric research.
There are two main goals of macroeconometrics:
Forecasting: Because we all want to learn about the future.
Studying the propagation of macroeconomic shocks: Because understanding how shocks to monetary policy, fiscal policy, uncertainty, trade, climate etc. will affect other variables is essential for policy.
About the course.
About the course.
Moodle
All materials will be posted in moodle:
http://dl1.cuni.cz/course/view.php?id=880
Enrollment key: wcmitchell
Schedule
1. Intro + Review of time series models (Lukáš Vácha); 15. 2.
2. Are univariate models relevant for forecasting? (Lukáš Vácha); 22.2.
3. Spectra, cycles, and filters. (Lukáš Vácha); 1.3.
4. Evaluating business cycles in real-time. Can we predict turning points? (Jaromír Baxa); 8.3.
5. State space and dynamic factor models: Synthesize many series for predictions. (Lukáš Vácha); 15.3.
6. A primer of Vector autoregressions. (Jaromír Baxa); 22.3.
7. Structural identification with VAR models (Jaromír Baxa); 29.3.
8. Identification via sign restrictions (Jaromír Baxa); 5.4.
9. Local projections + narrative identifications (Jaromír Baxa); 12.4.
10. Modelling nonstationary variables: Identification of long-run relationships. (Jaromír Baxa); 19.4.
11. Bayesian VARs: Employing a priori information and large datasets. (Jaromír Baxa); 26.4.
12. New approaches to structural identification: Proxy SVAR + high-frequency identification (Jaromír Baxa); 3.5.
(10.5. no lecture - rector’s day)
13. Nonlinearities + Spillovers (Jaromír Baxa); 17.5.
Course requirement: Final paper + Workshop presentation
Choose one country for which you will estimate the propagation of a shock of your interest and forecast the GDP growth and inflation.
The final paper will be prepared throughout the whole semester, with three intermediate deadlines:
after lecture 4: ARIMA model of inflation and GDP, estimation of potential structural breaks, estimation of spectra of both series, forecast 1 period and 1 year ahead; evaluation of a cyclical position of the economy
after lecture 8: VAR I - short-run restrictions, sign restrictions + VAR forecasts.
after lecture 11: VAR II - local projections, Bayesian VAR + BVAR forecasts.
The outcomes of these intermediate stages will be presented and discussed during the seminars.
Course requirement: Final paper + Workshop presentation
You are encouraged to work in pairs; however, each of you shall have his/her own estimates for a particular country.
The final paper can be written jointly, covering two countries as well.
Grades: Intermediate stages and presentations - 20 points each (60 in total), final paper + presentation, and participation at the workshop 40 points. The final paper and participation at the workshop are necessary conditions to pass the course, even if the sum of intermediate points exceeds 50.5 (which likely will).
Course requirement: Final paper + Workshop presentation
Reminder: In empirical projects, success and intuitive results are never assured. Experimentation is key. Therefore, for the final paper, it is more important and valuable to describe your modeling choice carefully, defend their appropriateness, and test the robustness of your results on these choices. Those choices include a selection of variables and their transformations, identifying assumptions, and model structure. It is the quality of research that makes any finding valuable, not the finding itself. It might easily happen to you that the results will not be in line with the mainstream intuition or that a simpler method will lead to more reasonable results than a more advanced model, but that’s okay.