Working Papers

2017-01 | October 2017


Measuring News Sentiment


We develop and assess new time series measures of economic sentiment based on computational text analysis of economic and financial newspaper articles from January 1980 to April 2015. The text analysis is based on predictive models estimated using machine learning techniques from Kanjoya. We analyze four alternative news sentiment indexes. The news sentiment indexes correlate strongly with contemporaneous business cycle indicators. Innovations to the indexes can predict future economic activity. We find that including the news sentiment indexes in a standard forecasting model can improve performance, in particular for inflation and the federal funds rate.

Download PDF (pdf, 1.59 mb)

Article Citation

Shapiro, Adam Hale, Moritz Sudhof, and Daniel Wilson. 2017. "Measuring News Sentiment," Federal Reserve Bank of San Francisco Working Paper 2017-01. Available at