Monday , December 23 2024
Home / SNB News / WP – 2024-06-04 – Milen Arro-Cannarsa and Rolf Scheufele: Nowcasting GDP: what are the gains from machine learning algorithms?

WP – 2024-06-04 – Milen Arro-Cannarsa and Rolf Scheufele: Nowcasting GDP: what are the gains from machine learning algorithms?

Summary:
We compare several machine learning methods for nowcasting GDP. A large mixed-frequency data set is used to investigate different algorithms such as regression based methods (LASSO, ridge, elastic net), regression trees (bagging, random forest, gradient boosting), and SVR. As benchmarks, we use univariate models, a simple forward selection algorithm, and a principal components regression. The analysis accounts for publication lags and treats monthly indicators as quarterly variables combined via blocking. Our data set consists of more than 1,100 time series. For the period after the Great Recession, which is particularly challenging in terms of nowcasting, we find that all considered machine learning techniques beat the univariate benchmark up to 28 % in terms of out-of-sample

Topics:
Swiss National Bank considers the following as important:

This could be interesting, too:

Dirk Niepelt writes “Report by the Parliamentary Investigation Committee on the Conduct of the Authorities in the Context of the Emergency Takeover of Credit Suisse”

Investec writes Federal parliament approves abolition of imputed rent

investrends.ch writes Novo Nordisk Studie bringt Absturz

Urs Birchler writes Der “Regulatorische Filter”: Das Filetstück des PUK-Berichts:

We compare several machine learning methods for nowcasting GDP. A large mixed-frequency data set is used to investigate different algorithms such as regression based methods (LASSO, ridge, elastic net), regression trees (bagging, random forest, gradient boosting), and SVR. As benchmarks, we use univariate models, a simple forward selection algorithm, and a principal components regression. The analysis accounts for publication lags and treats monthly indicators as quarterly variables combined via blocking. Our data set consists of more than 1,100 time series. For the period after the Great Recession, which is particularly challenging in terms of nowcasting, we find that all considered machine learning techniques beat the univariate benchmark up to 28 % in terms of out-of-sample RMSE. Ridge, elastic net, and SVR are the most promising algorithms in our analysis, significantly outperforming principal components regression.

Swiss National Bank
The Swiss National Bank conducts the country’s monetary policy as an independent central bank. It is obliged by the Constitution and by statute to act in accordance with the interests of the country as a whole. Its primary goal is to ensure price stability, while taking due account of economic developments. In so doing, it creates an appropriate environment for economic growth.

Leave a Reply

Your email address will not be published. Required fields are marked *