Download this Paper Open PDF in Browser

Structure in the Tweet Haystack: Uncovering the Link between Text-Based Sentiment Signals and Financial Markets

45 Pages Posted: 6 Sep 2015 Last revised: 8 Jan 2016

Axel Groß-Klußmann

Quoniam Asset Management GmbH

Markus Ebner

Quoniam Asset Management

Stephan König

Hannover University of Applied Sciences and Arts

Date Written: October 1, 2015

Abstract

We examine the relationship between signals derived from unstructured social media microblog text data and financial market developments. Employing statistical language modeling techniques we construct directional user sentiment and non-directional topic disagreement metrics and link these to S&P 500 index returns and volatility. Based on an extensive five year sample of Twitter messages our study shows that both unsupervised and supervised statistical learning methods successfully identify subsets of expert users with distinct finance focus. This allows to filter out the substantial noise associated with social media. Accounting for salient properties of the time series in ARMA models we document significant effects of expert disagreement signals on current and future S&P volatility. Moreover, we detect a significant contemporaneous relation between expert sentiment signals and S&P returns.

Keywords: Natural Language Processing, Sentiment Analysis, Unstructured Social Media Data, Big Data

JEL Classification: G14, C32

Suggested Citation

Groß-Klußmann, Axel and Ebner, Markus and König, Stephan, Structure in the Tweet Haystack: Uncovering the Link between Text-Based Sentiment Signals and Financial Markets (October 1, 2015). Available at SSRN: https://ssrn.com/abstract=2656204 or http://dx.doi.org/10.2139/ssrn.2656204

Axel Groß-Klußmann (Contact Author)

Quoniam Asset Management GmbH ( email )

Frankfurt
Germany

Markus Ebner

Quoniam Asset Management ( email )

Westhafenplatz 1
Frankfurt am Main, 60327
Germany

Stephan König

Hannover University of Applied Sciences and Arts ( email )

Ricklinger Stadtweg 120
Hannover, 30459
Germany

Paper statistics

Downloads
260
Rank
98,727
Abstract Views
1,598