Differential Effects of Buy and Sell Rules in Sentiment-Informed EUR/USD Trading

12 Pages Posted: 24 Feb 2017 Last revised: 27 Feb 2017

See all articles by Dietmar Janetzko

Dietmar Janetzko

Cologne Business School

Jonas Krauss

University of Cologne - Information Systems and Information Management

Stefan Nann

University of Cologne - Information Systems and Information Management

Date Written: July 15, 2016

Abstract

Sentiment indicators are often appealing to investors since financial markets are known to be influenced not only by economic fundamentals but also by emotions (Shiller & Akerlof, 2009). In current studies, however, a range of alternative sentiment indices is used each of which purports to have predictive value for the movements in financial markets. Consequently, a question mark remains in regards to the comparability of findings across different research papers. The goal of this study is to address the question of choice between competing sentiment indicators in EUR/USD trading. To identify the indicator having the best predictive value we estimate expected returns for individual sources and forecast models via backtesting (Campbell, 2005). Our findings support the notion that the predictive value depends on the source of the sentiment-indicator, on timing aspects, with more recent sentiments having greater predictive strength, and on the type of rule (e.g., buy/sell) harnessed.

Keywords: financial prediction, sentiment, sentiment analysis, social media, big data, machine learning, eurusd, stockpulse

Suggested Citation

Janetzko, Dietmar and Krauss, Jonas and Nann, Stefan, Differential Effects of Buy and Sell Rules in Sentiment-Informed EUR/USD Trading (July 15, 2016). Available at SSRN: https://ssrn.com/abstract=2922485 or http://dx.doi.org/10.2139/ssrn.2922485

Dietmar Janetzko

Cologne Business School ( email )

Hardefuststr. 1
Cologne, 50677
Germany

Jonas Krauss

University of Cologne - Information Systems and Information Management ( email )

Pohligstr. 1
Cologne, D-50969
Germany

Stefan Nann (Contact Author)

University of Cologne - Information Systems and Information Management ( email )

Pohligstr. 1
Cologne, D-50969
Germany

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

Paper statistics

Downloads
127
Abstract Views
1,022
rank
263,614
PlumX Metrics