Twitter mood predicts the stock market

Abstract

Abstract-Behavioral economics tells us that emotions can profoundly affect individual behavior and decision-making. Does this also apply to societies at large, i.e. can societies experience mood states that affect their collective decision making? By extension is the public mood correlated or even predictive of economic indicators? Here we investigate whether measurements of collective mood states derived from large-scale Twitter feeds are correlated to the value of the Dow Jones Industrial Average (DJIA) over time. We analyze the text content of daily Twitter feeds by two mood tracking tools, namely OpinionFinder that measures positive vs. negative mood and Google-Profile of Mood States (GPOMS) that measures mood in terms of 6 dimensions (Calm, Alert, Sure, Vital, Kind, and Happy). We cross-validate the resulting mood time series by comparing their ability to detect the public's response to the presidential election and Thanksgiving day in 2008. A Granger causality analysis and a Self-Organizing Fuzzy Neural Network are then used to investigate the hypothesis that public mood states, as measured by the OpinionFinder and GPOMS mood time series, are predictive of changes in DJIA closing values. Our results indicate that the accuracy of DJIA predictions can be significantly improved by the inclusion of specific public mood dimensions but not others. We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the DJIA and a reduction of the Mean Average Percentage Error by more than 6%.Index Terms-stock market prediction -twitter -mood analysis.

Financial Press and Stock Markets in Times of Crisis

“…They also found that disagreement between the posted messages was associated with increased trading volume. More recently, Bollen et al (2011) found that Twitter mood predicted more than 80% of daily volatility of closing values of the Dow Jones Industrial Average. This would confirm Nofsinger's argument that social mood may cause an increase of decisions biased by optimism or pessimism that can considerably influence aggregate investment and business activity, even reflecting future economic activities (Nofsinger (2005)).…”

“…Therefore, while our results cannot contribute to formulate a general theory of the link of press and markets, they can provide important clues to understand the 'social construction' of pessimism between press and markets. Among the limitations, it is important to note that press pessimism and market behaviour could be also conditioned by other media, such as the new social media and the Internet (see the recent study on Twitter by Bollen et al (2011)). Further investigation is needed to compare behaviour and impact of various media and provide a more precise analysis of the 2008/2009 crisis (e.g., see Preis et al (2010)).…”

“…While studies on the impact of social media on stock markets have been recently carried out that focused on similar periods of crisis (e.g., Preis et al (2010); Bollen et al (2011)), our idea was that, in a situation of financial turmoil, the authoritative columns of certain influential financial newspapers could even have a stronger impact on investors' mood. We opted to look at the front page banner headlines as they are crucial to summarize the meaning, tone and importance of the news but are not expected to contain true, precise and detailed information about economic facts extremely relevant for investors, unlike specialized columns.…”

Influence of Weak Labels for Emotion Recognition of Tweets

“…relationships and decision making and are therefore analysed in many research projects. The automatic detection of emotions in text allows for a broad range of applications, such as forecasting movie box office revenues during the opening weekend [14], 3D facial expression rendering based on recognized emotions in text [5], stock prediction [4], and helping to understand consumer views towards a product [3].…”

Affect Computation of Chinese Short Text

“…Such a rapid development of microblog industry has led to a great interest in sentiment analysis of the information published by the microblog users [1]. In fact, tentative studies have already been conducted to analyze the sentiment of messages in Twitter, blogs and Google search queries to predict stock market, product sales, consumer spending, and movie sales [2]- [7].…”