Category: Blogs
-
Retooling Our Models
Some econometric models “broke” during the Covid era. Notably, the New York Fed’s GDP Nowcasting model was suspended. Basically, data were so volatile that their model was no longer delivering good predictions. The natural question is how will it be re-tooled? How will so many of the models out there be re-tooled? For macro forecasters, […]
-
How Machine Learning Can Save Economics
Bad model predictions and incredible model assumptions invite vitriol upon economists. Luckily, measure critique can lead to better economic models. The demise of large Keynesian simultaneous-equation models ultimately yielded a vibrant literature producing VARs, DSGEs, dynamic factor models, and more. Models ought to always be improving. And many see that machine learning (ML) has the […]
-
Forecasting With Level VARs Despite Non-Stationarity And/Or Cointegration. Intuition.
I sometimes run into people who are violently opposed to fitting VARs to non-stationary or cointegrated data. For sure, there are problems with frequentist inference and IRFs under these conditions. See Sims, Stock, and Watson (1990). But forecasting ability is not necessarily destroyed. Here, I am talking about vanilla OLS VARs, to say nothing of […]
-
Macro Random Forest Leads To Macro Gains In Macro Forecasting
Tomorrow is the 11th ECB Forecasting Conference. I am excited to see so many top authors: Sims, Engle, Koop, Marcellino, Schorfeide, and many more. It is fitting that so many progenitors of the innovative models of yesteryear and workhorse methods of today — VAR, ARCH, and Bayesian macroeconometrics — are here to oversee the next […]