Using Social Media to Identify Market Inefficiencies: Evidence from Twitter and Betfair

Alasdair Brown; Dooruj Rambaccussing; J. James  Reade; Giambattista  Rossi

Using Social Media to Identify Market Inefficiencies: Evidence from Twitter and Betfair

Alasdair Brown (Lead / Corresponding author), Dooruj Rambaccussing, J. James Reade, Giambattista Rossi

Research output: Working paper

Abstract

Information extracted from social media has been used by academics, and increasingly
by practitioners, to predict stock returns. But to what extent does social media
output predict asset fundamentals, and not simply short-term returns? In this paper we
analyse 13.8m posts on Twitter, and high-frequency betting data from Betfair, concerning
English Premier League soccer matches in 2013/14. Crucially, asset fundamentals
are revealed at the end of play. We find that the Tweets of certain journalists, and the
tone of all Tweets, contain fundamental information not revealed in betting prices. In
particular, Tweets aid in the interpretation of news during matches.

Original language	English
Place of Publication	Dundee
Publisher	University of Dundee, Department of Economic Studies
Number of pages	32
Publication status	Published - 17 Feb 2016

Publication series

Name	Dundee Discussion Papers in Economics
Publisher	University of Dundee, Department of Economic Studies
No.	293
ISSN (Print)	1473-236X

Keywords

Social media
Prediction markets
Fundamentals
Sentiment
Mispricing

Access to Document

Link to item via University of Dundee website

Fingerprint Dive into the research topics of 'Using Social Media to Identify Market Inefficiencies: Evidence from Twitter and Betfair'. Together they form a unique fingerprint.

Cite this

Brown, A., Rambaccussing, D., Reade, J. J., & Rossi, G. (2016). Using Social Media to Identify Market Inefficiencies: Evidence from Twitter and Betfair. (Dundee Discussion Papers in Economics; No. 293). University of Dundee, Department of Economic Studies .

@techreport{290d787e0b204ca9bcfa827689c2283f,

title = "Using Social Media to Identify Market Inefficiencies: Evidence from Twitter and Betfair",

abstract = "Information extracted from social media has been used by academics, and increasinglyby practitioners, to predict stock returns. But to what extent does social mediaoutput predict asset fundamentals, and not simply short-term returns? In this paper weanalyse 13.8m posts on Twitter, and high-frequency betting data from Betfair, concerningEnglish Premier League soccer matches in 2013/14. Crucially, asset fundamentalsare revealed at the end of play. We find that the Tweets of certain journalists, and thetone of all Tweets, contain fundamental information not revealed in betting prices. Inparticular, Tweets aid in the interpretation of news during matches.",

keywords = "Social media, Prediction markets, Fundamentals, Sentiment, Mispricing",

author = "Alasdair Brown and Dooruj Rambaccussing and Reade, {J. James} and Giambattista Rossi",

note = "Alasdair Brown acknowledges financial support from the British Academy and the Leverhulme Trust (Ref:SG140097), Dooruj Rambaccussing acknowledges financial support from an SIRE Early Career Grant, James Reade thanks the Open Society Foundations (OSF) and Oxford Martin School (OMS) for financial support, and Dooruj Rambaccussing, James Reade and Giambattista Rossi acknowledge support from Twitter through their #DataGrants scheme. The research presented in this paper was carried out on the High Performance Computing Cluster supported by the Research and Specialist Computing Support service at the University of East Anglia. The usual disclaimer applies",

year = "2016",

month = feb,

day = "17",

language = "English",

series = "Dundee Discussion Papers in Economics",

publisher = "University of Dundee, Department of Economic Studies ",

number = "293",

type = "WorkingPaper",

institution = "University of Dundee, Department of Economic Studies ",

}

TY - UNPB

T1 - Using Social Media to Identify Market Inefficiencies

T2 - Evidence from Twitter and Betfair

AU - Brown, Alasdair

AU - Rambaccussing, Dooruj

AU - Reade, J. James

AU - Rossi, Giambattista

N1 - Alasdair Brown acknowledges financial support from the British Academy and the Leverhulme Trust (Ref:SG140097), Dooruj Rambaccussing acknowledges financial support from an SIRE Early Career Grant, James Reade thanks the Open Society Foundations (OSF) and Oxford Martin School (OMS) for financial support, and Dooruj Rambaccussing, James Reade and Giambattista Rossi acknowledge support from Twitter through their #DataGrants scheme. The research presented in this paper was carried out on the High Performance Computing Cluster supported by the Research and Specialist Computing Support service at the University of East Anglia. The usual disclaimer applies

PY - 2016/2/17

Y1 - 2016/2/17

N2 - Information extracted from social media has been used by academics, and increasinglyby practitioners, to predict stock returns. But to what extent does social mediaoutput predict asset fundamentals, and not simply short-term returns? In this paper weanalyse 13.8m posts on Twitter, and high-frequency betting data from Betfair, concerningEnglish Premier League soccer matches in 2013/14. Crucially, asset fundamentalsare revealed at the end of play. We find that the Tweets of certain journalists, and thetone of all Tweets, contain fundamental information not revealed in betting prices. Inparticular, Tweets aid in the interpretation of news during matches.

AB - Information extracted from social media has been used by academics, and increasinglyby practitioners, to predict stock returns. But to what extent does social mediaoutput predict asset fundamentals, and not simply short-term returns? In this paper weanalyse 13.8m posts on Twitter, and high-frequency betting data from Betfair, concerningEnglish Premier League soccer matches in 2013/14. Crucially, asset fundamentalsare revealed at the end of play. We find that the Tweets of certain journalists, and thetone of all Tweets, contain fundamental information not revealed in betting prices. Inparticular, Tweets aid in the interpretation of news during matches.

KW - Social media

KW - Prediction markets

KW - Fundamentals

KW - Sentiment

KW - Mispricing

UR - https://www.dundee.ac.uk/business/research/discussionpapersseries/

M3 - Working paper

T3 - Dundee Discussion Papers in Economics

BT - Using Social Media to Identify Market Inefficiencies

PB - University of Dundee, Department of Economic Studies

CY - Dundee

ER -