From 2018 to 2020, I taught Applied Macroeconometrics at Pontificia Universidad Católica de Chile’s M.A. Program in Applied Economics.
Here is the most recent syllabus of the course (in Spanish), which I taught jointly with my colleague Agustín Arias.
I share below slides, applications and code on selected topics. This material is in Spanish, except the code, which is heavily commented in English. Note that the purpose of the code is pedagogy, not efficiency. I borrow heavily from Andrew Blake and Haroon Mumtaz’s excellent handbook “Applied Bayesian Econometrics for Central Bankers.” If you find any errors, please let me know (jguerra@bcentral.cl).
Reduced-form vector autoregressions and forecasting
[Slides] | [Data and code]
In addition to background material, the slides include the descrption and results of an application to inflation forecast accuracy in Chile. The application compares the forecasting performance of: i) univariate autoregressive models, and ii) VARs that include the nominal exchange rate in addition to inflation. The data and code replicate these results.
Structural vector autoregressions and causality
Bayesian vector autoregressions
[Slides] | [Data and code]
In addition to background material, the slides include the descrption and results of two applications to monetary policy in Chile. Using the same dataset as in the previous topic, there is material for Bayesian estimation with two structural identification strategies: i) the first assumes a recursive structure and conducts a Cholesky factorization of the variance-covariance matrix of reduced-form residuals; whereas ii) the second imposes restrictions on the signs of responses to monetary policy shocks. The data and code replicate these results.