News Media vs. FRED-MD for Macroeconomic Forecasting

45 Pages Posted: 28 Oct 2020

See all articles by Jon Ellingsen

Jon Ellingsen

BI Norwegian Business School

Vegard H. Larsen

BI Norwegian Business School, Department of Data Science and Analytics

Leif Anders Thorsrud

Centre for Applied Macroeconomics and Commodity Prices (CAMP), BI Norwegian Business School

Date Written: 2020

Abstract

Using a unique dataset of 22.5 million news articles from the Dow Jones Newswires Archive, we perform an in depth real-time out-of-sample forecasting comparison study with one of the most widely used data sets in the newer forecasting literature, namely the FRED-MD dataset. Focusing on U.S. GDP, consumption and investment growth, our results suggest that the news data contains information not captured by the hard economic indicators, and that the news-based data are particularly informative for forecasting consumption developments.

Keywords: forecasting, real-time, machine learning, news, text data

JEL Classification: C530, C550, E270, E370

Suggested Citation

Ellingsen, Jon and Larsen, Vegard H. and Thorsrud, Leif Anders, News Media vs. FRED-MD for Macroeconomic Forecasting (2020). CESifo Working Paper No. 8639, Available at SSRN: https://ssrn.com/abstract=3720396 or http://dx.doi.org/10.2139/ssrn.3720396

Jon Ellingsen (Contact Author)

BI Norwegian Business School ( email )

Nydalsveien 37
Oslo, 0442
Norway

Vegard H. Larsen

BI Norwegian Business School, Department of Data Science and Analytics ( email )

Nydalsveien 37
Oslo, 0442
Norway

Leif Anders Thorsrud

Centre for Applied Macroeconomics and Commodity Prices (CAMP), BI Norwegian Business School ( email )

Nydalsveien 37
Oslo, 0442
Norway

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
160
Abstract Views
631
Rank
393,722
PlumX Metrics