News Beta: Factoring Sentiment Risk into Quant Models

Posted: 1 Jun 2012 Last revised: 10 Mar 2017

Date Written: February 2, 2012

Abstract

The stock market is affected by sentiment. The question is, however, how to quantify this effect on asset prices. By utilizing the unique RavenPack Sentiment Index, a news-based proxy for market sentiment, this paper intends to address this issue empirically by exploring the pricing implications of a stock’s exposure to market sentiment. We also explore a concept we coined as "news beta" or the sensitivity of stock returns to changes in market sentiment as reported by the media. After controlling for traditional factors, news beta is found to have strong return predictability over 6 and 12 month horizons. The evidence from this research suggests that market sentiment data is still an untapped source of alpha in financial markets.

Keywords: market sentiment, news analytics, news beta, factor models, news sentiment, sentiment, stock selection

JEL Classification: G1

Suggested Citation

Hafez, Peter and Xie, Junqiang, News Beta: Factoring Sentiment Risk into Quant Models (February 2, 2012). Journal of Investing, Vol. 25, No. 3, 2016. Available at SSRN: https://ssrn.com/abstract=2071142 or http://dx.doi.org/10.2139/ssrn.2071142

Peter Hafez (Contact Author)

RavenPack ( email )

260 Madison Ave., 8th Floor
New York, NY 10016
United States
+34 662 180 797 (Phone)

HOME PAGE: http://www.ravenpack.com

Junqiang Xie

RavenPack ( email )

260 Madison Ave., 8th Floor
New York, NY 10016
United States
+1 646 216 2140 (Phone)

HOME PAGE: http://www.ravenpack.com

Register to save articles to
your library

Register

Paper statistics

Abstract Views
3,361
PlumX Metrics