Time Series Econometrics, second half
Economics 513
Spring 2021

Taught by: Chris Sims

Contact information


Syllabus
Link to last previous version of the course
New information is in red.



The course meets 10:40-12:10 Tuesday and Thursday on zoom .

Exercises, links to additional notes, announcements about the course, etc. will appear here.

Link to last version of ECO517
This course included a discussion of Bayesian inference for the standard normal linear model (SNLM), which we'll assume known in 513. You can look first at Lecture 9, slides 10-12. These draw on the discussion of inference for the normal distribution in Lecture 6b. If you have not had much previous exposure to the Bayesian framework for inference, scanning through earlier lectures in that course could also be helpful.
Hidden Chains notes
BPEA Phillips Curve discussion
Dynamic factor models notes
Model Comparison notes
Exercise on IDH, model comparison
Notes on error bands for impulse responses
Testing restrictions.
These notes overlap with the material in the Model Comparison notes, but put more emphasis on the case of zero restrictions on linear regressions. They were not the basis of any lecture this year.
Minnesota prior
Scaling in calculating Bayes factors
Cointegration
Seasonality
Seasonality Exercise
Particle Filter
Continuous time models
Note that the lecture on this topic went somewhat beyond these slides.
2018 general exam question from 513, with answer<
Notes on structural VAR's and system likelihood from the 2019 version of the coursemight be useful to you. They in part overlap with material from the first half of this year's 513, but they also include a lot of material we did not cover and that you are not responsible for. The SVAR notes on identification and on the likelihood factorization might be the most useful.
Final exam You can spend up to 2 hours on the exam, not counting time for reformatting and uploading. It should be returned by 9:30AM Friday, 4/30.