Evaluating sentiment in financial news articles

Robert P. Schumaker, Yulei Zhang, Chun Neng Huang, Hsinchun Chen

Research output: Contribution to journalArticle

117 Citations (Scopus)

Abstract

Can the choice of words and tone used by the authors of financial news articles correlate to measurable stock price movements? If so, can the magnitude of price movement be predicted using these same variables? We investigate these questions using the Arizona Financial Text (AZFinText) system, a financial news article prediction system, and pair it with a sentiment analysis tool. Through our analysis, we found that subjective news articles were easier to predict in price direction (59.0% versus 50.0% of chance alone) and using a simple trading engine, subjective articles garnered a 3.30% return. Looking further into the role of author tone in financial news articles, we found that articles with a negative sentiment were easiest to predict in price direction (50.9% versus 50.0% of chance alone) and a 3.04% trading return. Investigating negative sentiment further, we found that our system was able to predict price decreases in articles of a positive sentiment 53.5% of the time, and price increases in articles of a negative sentiment 52.4% of the time. We believe that perhaps this result can be attributable to market traders behaving in a contrarian manner, e.g., see good news, sell; see bad news, buy.

Original languageEnglish (US)
Pages (from-to)458-464
Number of pages7
JournalDecision Support Systems
Volume53
Issue number3
DOIs
StatePublished - Jun 2012

Fingerprint

Engines
News Articles
Sentiment
News
Direction compound

Keywords

  • Business intelligence
  • Financial prediction
  • Sentiment analysis
  • Text mining

ASJC Scopus subject areas

  • Management Information Systems
  • Information Systems
  • Information Systems and Management
  • Arts and Humanities (miscellaneous)
  • Developmental and Educational Psychology

Cite this

Evaluating sentiment in financial news articles. / Schumaker, Robert P.; Zhang, Yulei; Huang, Chun Neng; Chen, Hsinchun.

In: Decision Support Systems, Vol. 53, No. 3, 06.2012, p. 458-464.

Research output: Contribution to journalArticle

Schumaker, Robert P. ; Zhang, Yulei ; Huang, Chun Neng ; Chen, Hsinchun. / Evaluating sentiment in financial news articles. In: Decision Support Systems. 2012 ; Vol. 53, No. 3. pp. 458-464.
@article{e98ec1711a274278b32679ae3f451935,
title = "Evaluating sentiment in financial news articles",
abstract = "Can the choice of words and tone used by the authors of financial news articles correlate to measurable stock price movements? If so, can the magnitude of price movement be predicted using these same variables? We investigate these questions using the Arizona Financial Text (AZFinText) system, a financial news article prediction system, and pair it with a sentiment analysis tool. Through our analysis, we found that subjective news articles were easier to predict in price direction (59.0{\%} versus 50.0{\%} of chance alone) and using a simple trading engine, subjective articles garnered a 3.30{\%} return. Looking further into the role of author tone in financial news articles, we found that articles with a negative sentiment were easiest to predict in price direction (50.9{\%} versus 50.0{\%} of chance alone) and a 3.04{\%} trading return. Investigating negative sentiment further, we found that our system was able to predict price decreases in articles of a positive sentiment 53.5{\%} of the time, and price increases in articles of a negative sentiment 52.4{\%} of the time. We believe that perhaps this result can be attributable to market traders behaving in a contrarian manner, e.g., see good news, sell; see bad news, buy.",
keywords = "Business intelligence, Financial prediction, Sentiment analysis, Text mining",
author = "Schumaker, {Robert P.} and Yulei Zhang and Huang, {Chun Neng} and Hsinchun Chen",
year = "2012",
month = "6",
doi = "10.1016/j.dss.2012.03.001",
language = "English (US)",
volume = "53",
pages = "458--464",
journal = "Decision Support Systems",
issn = "0167-9236",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Evaluating sentiment in financial news articles

AU - Schumaker, Robert P.

AU - Zhang, Yulei

AU - Huang, Chun Neng

AU - Chen, Hsinchun

PY - 2012/6

Y1 - 2012/6

N2 - Can the choice of words and tone used by the authors of financial news articles correlate to measurable stock price movements? If so, can the magnitude of price movement be predicted using these same variables? We investigate these questions using the Arizona Financial Text (AZFinText) system, a financial news article prediction system, and pair it with a sentiment analysis tool. Through our analysis, we found that subjective news articles were easier to predict in price direction (59.0% versus 50.0% of chance alone) and using a simple trading engine, subjective articles garnered a 3.30% return. Looking further into the role of author tone in financial news articles, we found that articles with a negative sentiment were easiest to predict in price direction (50.9% versus 50.0% of chance alone) and a 3.04% trading return. Investigating negative sentiment further, we found that our system was able to predict price decreases in articles of a positive sentiment 53.5% of the time, and price increases in articles of a negative sentiment 52.4% of the time. We believe that perhaps this result can be attributable to market traders behaving in a contrarian manner, e.g., see good news, sell; see bad news, buy.

AB - Can the choice of words and tone used by the authors of financial news articles correlate to measurable stock price movements? If so, can the magnitude of price movement be predicted using these same variables? We investigate these questions using the Arizona Financial Text (AZFinText) system, a financial news article prediction system, and pair it with a sentiment analysis tool. Through our analysis, we found that subjective news articles were easier to predict in price direction (59.0% versus 50.0% of chance alone) and using a simple trading engine, subjective articles garnered a 3.30% return. Looking further into the role of author tone in financial news articles, we found that articles with a negative sentiment were easiest to predict in price direction (50.9% versus 50.0% of chance alone) and a 3.04% trading return. Investigating negative sentiment further, we found that our system was able to predict price decreases in articles of a positive sentiment 53.5% of the time, and price increases in articles of a negative sentiment 52.4% of the time. We believe that perhaps this result can be attributable to market traders behaving in a contrarian manner, e.g., see good news, sell; see bad news, buy.

KW - Business intelligence

KW - Financial prediction

KW - Sentiment analysis

KW - Text mining

UR - http://www.scopus.com/inward/record.url?scp=84862521235&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862521235&partnerID=8YFLogxK

U2 - 10.1016/j.dss.2012.03.001

DO - 10.1016/j.dss.2012.03.001

M3 - Article

VL - 53

SP - 458

EP - 464

JO - Decision Support Systems

JF - Decision Support Systems

SN - 0167-9236

IS - 3

ER -