Sentiment Derived from News Predicts EURUSD Movements

12 Pages Posted: 19 Oct 2013

Date Written: June 11, 2013

Abstract

In this research, we investigate how EURUSD reacts to short and long term changes in macro sentiment between the U.S. and the Eurozone. Sentiment is measured by assigning scores to macroeconomic and geopolitical events reported in the news. Overall, we find evidence that the EURUSD overreacts to daily changes in macro sentiment but underreacts to monthly changes. This phenomenon leads to a sentiment momentum effect. A contrarian strategy with a one-day holding period on EURUSD generates an average return of 7.2% per year with an Information Ratio of 0.75 after transaction costs. A momentum strategy with a 1-month holding period generates an average return of 3.3% per year with an Information Ratio of 1.60 after transaction costs.

Keywords: EURUSD, RavenPack, news analytics, sentiment, reversal, momentum, macro, Forex, FX

JEL Classification: F31

Suggested Citation

Hafez, Peter and Xie, Junqiang, Sentiment Derived from News Predicts EURUSD Movements (June 11, 2013). Available at SSRN: https://ssrn.com/abstract=2342247 or http://dx.doi.org/10.2139/ssrn.2342247

Peter Hafez (Contact Author)

RavenPack International ( email )

Centro de Negocios Oasis, Of. 8
Ctra. de Cadiz Km. 176
Marbella, Malaga 29602
Spain
+44 (0)20 3290 7395 (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

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
425
Abstract Views
2,576
Rank
127,199
PlumX Metrics