← Search

Journal of Financial and Quantitative Analysis Vol. 54 No. 5 2019

Does Unusual News Forecast Market Stress?

Paul Glasserman; Harry Mamaysky

Columbia University

Abstract

An increase in “unusual” news with negative sentiment predicts an increase in stock market volatility. Unusual positive news forecasts lower volatility. Our analysis is based on over 360,000 articles on 50 large financial companies, published during the period of 1996–2014. Unusualness interacted with sentiment forecasts company-specific and aggregate volatility several months ahead. Furthermore, unusual news is reflected more slowly in aggregate volatility than company-specific volatility. News measures from articles explicitly about the “market,” which are more easily accessible to investors, do not forecast volatility. The observed responses of volatility to news may be explained by attention constraints on investors.

DOI
10.1017/s0022109019000127
Volume
54
Issue
5
Pages
1937-1974
Language
en
Sources
bibtex:phds-export.bib openalex crossref

Cite