Please find below a list of selected publication related to my research topic. You can filter by type to
Social media are increasingly reflecting and influencing behavior of other complex systems. In this paper we investigate the relations between a well-known micro-blogging platform Twitter and financial markets. In particular, we consider, in a period of 15 months, the Twitter volume and sentiment about the 30 stock companies that form the Dow Jones Industrial Average (DJIA) index. We find a relatively low Pearson correlation and Granger causality between the corresponding time series over the entire time period. However, we find a significant dependence between the Twitter sentiment and abnormal returns during the peaks of Twitter volume. This is valid not only for the expected Twitter volume peaks (e.g., quarterly announcements), but also for peaks corresponding to less obvious events. We formalize the procedure by adapting the well-known event study from economics and finance to the analysis of Twitter data. The procedure allows to automatically identify events as Twitter volume peaks, to compute the prevailing sentiment (positive or negative) expressed in tweets at these peaks, and finally to apply the event study methodology to relate them to stock returns. We show that sentiment polarity of Twitter peaks implies the direction of cumulative abnormal returns. The amount of cumulative abnormal returns is relatively low (about 12%), but the dependence is statistically significant for several days after the events.
We examine the impact of more than 2.5 million HotCopper messages on the Australian stock market. HotCopper is the largest online stock message board in Australia and the sample of messages covers over 2000 companies listed on the Australian Securities Exchange (ASX) from January 2003 through December 2008. We exclude messages surrounding public price-sensitive announcements released centrally by the ASX in order to examine the private information content of internet board messages. We find that the number of board messages and message sentiment significantly and positively relate to the contemporaneous returns of underperforming (low ROE, EBIT margin, EPS) small capitalization stocks with high market growth potential (low book-to-market). Posting activity is positively associated with trading volume for small stocks and negatively associated with bid-ask spreads for small and large stocks in the short term. Bullish small stocks outperform bearish ones significantly in respective days and months, exhibiting no return reversals to pre-message board activity levels in subsequent time periods. Large stocks are not found to be affected by message board activity. We conclude that higher message board activity quickly reflects itself into the prices of small capitalization stocks in a highly regulated market like the ASX.
We use daily Internet search volume from millions of households to reveal market-level sentiment. By aggregating the volume of queries related to household concerns (e.g., recession," unemployment," and bankruptcy"), we construct a Financial and Economic Attitudes Revealed by Search (FEARS) index as a new measure of investor sentiment. Between 2004 and 2011, we find FEARS (i) predict short-term return reversals, (ii) predict temporary increases in volatility, and (iii) predict mutual fund flows out of equity funds and into bond funds. Taken together, the results are broadly consistent with theories of investor sentiment.
We study approximately 5.0 million stock picks submitted by individual users to the CAPS website run by the Motley Fool company (www.caps.fool.com). These picks prove to be surprisingly informative about future stock prices. Shorting stocks with a disproportionate number of negative picks and buying stocks with a disproportionate number of positive picks yields a return of over 12% per annum over the sample period. Negative picks mostly drive these results; they strongly predict future stock price declines. Returns to positive picks are statistically indistinguishable from the market. A FamaFrench decomposition suggests that stock-picking rather than style factors largely produced these results.
A blindfolded chimpanzee throwing darts at The Wall Street Journal could select a portfolio that would do as well as the (stock market) experts [Malkiel (2003) The efficient market hypothesis and its critics. Journal of Economic Perspectives 17(1): 5982)]. However, what if this chimpanzee could browse the Internet before throwing any darts? In this paper, we ask whether online news has any influence on the financial market, and we also investigate how much influence it has. We explore the burgeoning literature on the predictability of financial movements using online information and report its mixed findings. In addition, we collate the efforts of various disciplines, including economics, text mining, sentiment analysis and machine learning, and we offer suggestions for future research.
We examine the relation between daily sentiment and trading behavior within 20 international markets by exploiting Facebook's Gross National Happiness Index. We find that sentiment has a positive contemporaneous relation to stock returns. Moreover, sentiment on Sunday affects stock returns on Monday, suggesting causality from sentiment to stock markets. We observe that the relation between sentiment and returns reverses the following weeks. We further show that negative sentiments are related to increases in trading volume and return volatility. These results highlight the importance of behavioral factors in stock investing.
By using an extensive dataset of more than 32 million messages on 91 firms posted on the Yahoo! Finance message board over the period January 2005 to December 2010, we examine whether investor sentiment as expressed in posted messages has predictive power for stock returns, volatility, and trading volume. In intertemporal and cross-sectional regression analyses, we find no evidence that investor sentiment forecasts future stock returns either at the aggregate or at the individual firm level. Rather, we find evidence that investor sentiment is positively affected by prior stock price performance. We also find no significant evidence that investor sentiment from Internet postings has predictive power for volatility and trading volume. A distinctive feature of our study is the use of sentiment information explicitly revealed by retail investors as well as classified by a machine learning classification algorithm and a much longer sample period relative to prior studies.
Recent studies in behavioral finance discover that emotional impulses of stock investors affect stock prices. The challenge lies in how to quantify such sentiment to predict stock market movements. In this article, we propose a media-aware quantitative trading strategy utilizing sentiment information of Web media. This is achieved by capturing public mood from interactive behaviors of investors in social media and studying the impact of firm-specific news sentiment on stocks along with such public mood. Our experiments on the CSI 100 stocks during a three-month period show that a predictive performance in closeness to the actual future stock price is 0.612 in terms of root mean squared error, the same direction of price movement as the future price is 55.08%, and a simulation trading return is up to 166.11%.
This study presents a methodology for identifying a broad range of real-world news events based on microblogging messages. Applying computational linguistics to a unique dataset of more than 400,000 S&P 500 stock-related Twitter messages, we distinguish between good and bad news and demonstrate that the returns prior to good news events are more pronounced than for bad news events. We show that the stock market impact of news events differs substantially across different categories.
We survey the textual sentiment literature, comparing and contrasting the various information sources, content analysis methods, and empirical models that have been used to date. We summarize the important and influential findings about how textual sentiment impacts on individual, firm-level and market-level behavior and performance, and vice versa. We point to what is agreed and what remains controversial. Promising directions for future research are emerging from the availability of more accurate and efficient sentiment measures resulting from increasingly sophisticated textual content analysis coupled with more extensive field-specific dictionaries. This is enabling more wide-ranging studies that use increasingly sophisticated models to help us better understand behavioral finance patterns across individuals, institutions and markets.
In this paper, we examine the intra-day effects of verbal statements and comments on the FX market uncertainty using two measures: continuous volatility and discontinuous jumps. Focusing on the euro-dollar exchange rate, we provide empirical evidence of how these two sources of uncertainty matter in measuring the short-term reaction of exchange rates to communication events. Talks significantly trigger large jumps or extreme events for approximately an hour after the news release. Continuous volatility starts reacting prior to the news, intensifies around the release time and stays at high levels for several hours. Our results suggest that monetary authorities generally tend to communicate with markets on days when uncertainty is relatively severe, and higher than normal. Disentangling the US and Euro area statements, we also find that abnormal levels of volatility are mostly driven by the communication of the Euro area officials rather than US authorities.
Microblogging forums (e.g., Twitter) have become a vibrant online platform for exchanging stock-related information. Using methods from computational linguistics, we analyse roughly 250,000 stock-related messages (so-called tweets) on a daily basis. We find an association between tweet sentiment and stock returns, message volume and trading volume, as well as disagreement and volatility. In contrast to previous related research, we also analyse the mechanism leading to an efficient aggregation of information in microblogging forums. Our results demonstrate that users providing above average investment advice are retweeted (i.e., quoted) more often and have more followers, which amplifies their share of voice.
Social media has become a popular venue for individuals to share the results of their own analysis on financial securities. This paper investigates the extent to which investor opinions transmitted through social media predict future stock returns and earnings surprises. We conduct textual analysis of articles published on one of the most popular social media platforms for investors in the United States. We also consider the readers' perspective as inferred via commentaries written in response to these articles. We find that the views expressed in both articles and commentaries predict future stock returns and earnings surprises.
This paper studies the effect of sentiment on asset prices during the 20th century (1905 to 2005). As a proxy for sentiment, we use the fraction of positive and negative words in two columns of financial news from the New York Times. The main contribution of the paper is to show that, controlling for other well-known time-series patterns, the predictability of stock returns using news' content is concentrated in recessions. A one standard deviation shock to our news measure during recessions predicts a change in the conditional average return on the DJIA of 12 basis points over one day.
We present a new approach for content analysis to quantify document tone. We find a significant relation between our measure of the tone of 10-Ks and market reaction for both negative and positive words. We also find that the appropriate choice of term weighting in content analysis is at least as important as, and perhaps more important than, a complete and accurate compilation of the word list. Furthermore, we show that our approach circumvents the need to subjectively partition words into positive and negative word lists. Our approach reliably quantifies the tone of IPO prospectuses as well, and we find that the document score is negatively related to IPO underpricing.
We use exogenous scheduling of Wall Street Journal columnists to identify a causal relation between financial reporting and stock market performance. To measure the media's unconditional effect, we add columnist fixed effects to a daily regression of excess Dow Jones Industrial Average returns. Relative to standard control variables, these fixed effects increase the R2 by about 35%, indicating each columnist's average persistent bullishness or bearishness. To measure the media's conditional effect, we interact columnist fixed effects with lagged returns. This increases explanatory power by yet another one-third, and identifies amplification or attenuation of prevailing sentiment as a tool used by financial journalists.
We propose a new and direct measure of investor attention using search frequency in Google (Search Volume Index (SVI)). In a sample of Russell 3000 stocks from 2004 to 2008, we find that SVI (1) is correlated with but different from existing proxies of investor attention; (2) captures investor attention in a more timely fashion and (3) likely measures the attention of retail investors. An increase in SVI predicts higher stock prices in the next 2 weeks and an eventual price reversal within the year. It also contributes to the large first-day return and long-run underperformance of IPO stocks.
Opening, lunch and closing of financial markets induce a periodic component in the volatility of high-frequency returns. We show that price jumps cause a large bias in the classical periodicity estimators and propose robust alternatives. We find that accounting for periodicity greatly improves the accuracy of intraday jump detection methods. It increases the power to detect the relatively small jumps occurring at times for which volatility is periodically low and reduces the number of spurious jump detections at times of periodically high volatility. We use the series of detected jumps to estimate robustly the long memory parameter of the squared EUR/USD, GBP/USD and YEN/USD returns.
Previous research uses negative word counts to measure the tone of a text. We show that word lists developed for other disciplines misclassify common words in financial text. In a large sample of 10-Ks during 1994 to 2008, almost three-fourths of the words identified as negative by the widely used Harvard Dictionary are words typically not considered negative in financial contexts. We develop an alternative negative word list, along with five other word lists, that better reflect tone in financial text. We link the word lists to 10-K filing returns, trading volume, return volatility, fraud, material weakness, and unexpected earnings.
We test and confirm the hypothesis that individual investors are net buyers of attention-grabbing stocks, e.g., stocks in the news, stocks experiencing high abnormal trading volume, and stocks with extreme one-day returns. Attention-driven buying results from the difficulty that investors have searching the thousands of stocks they can potentially buy. Individual investors do not face the same search problem when selling because they tend to sell only stocks they already own. We hypothesize that many investors consider purchasing only stocks that have first caught their attention. Thus, preferences determine choices after attention has determined the choice set.
We examine whether a simple quantitative measure of language can be used to predict individual firms accounting earnings and stock returns. Our three main findings are: (1) the fraction of negative words in firm-specific news stories forecasts low firm earnings; (2) firms stock prices briefly underreact to the information embedded in negative words; and (3) the earnings and return predictability from negative words is largest for the stories that focus on fundamentals. Together these findings suggest that linguistic media content captures otherwise hard-to-quantify aspects of firms fundamentals, which investors quickly incorporate into stock prices.
I quantitatively measure the interactions between the media and the stock market using daily content from a popularWall Street Journal column. I find that high media pessimism predicts downward pressure on market prices followed by a reversion to fundamentals, and unusually high or low pessimism predicts high market trading volume. These and similar results are consistent with theoretical models of noise and liquidity traders, and are inconsistent with theories of media content as a proxy for new information about fundamental asset values, as a proxy for market volatility, or as a sideshow with no relationship to asset markets.
Real investors and markets are too complicated to be neatly summarized by a few selected biases and trading frictions. The "top down" approach to behavioral finance focuses on the measurement of reduced form, aggregate sentiment and traces its effects to stock returns. It builds on the two broader and more irrefutable assumptions of behavioral finance -- sentiment and the limits to arbitrage -- to explain which stocks are likely to be most affected by sentiment. In particular, stocks of low capitalization, younger, unprofitable, high volatility, non-dividend paying, growth companies, or stocks of firms in financial distress, are likely to be disproportionately sensitive to broad waves of investor sentiment. We review the theoretical and empirical evidence for these predictions.
Extracting sentiment from text is a hard semantic problem. We develop a methodology for extracting small investor sentiment from stock message boards. The algorithm comprises different classifier algorithms coupled together by a voting scheme. Accuracy levels are similar to widely used Bayes classifiers, but false positives are lower and sentiment accuracy higher. Time series and cross-sectional aggregation of message information improves the quality of the resultant sentiment index, particularly in the presence of slang and ambiguity. Empirical applications evidence a relationship with stock valuestech-sector postings are related to stock index levels, and to volumes and volatility. The algorithms may be used to assess the impact on investor opinion of management announcements, press releases, third-party news, and regulatory changes.
We study how investor sentiment affects the cross-section of stock returns. We predict that a wave of investor sentiment has larger effects on securities whose valuations are highly subjective and difficult to arbitrage. Consistent with this prediction, we find that when beginning-of-period proxies for sentiment are low, subsequent returns are relatively high for small stocks, young stocks, high volatility stocks, unprofitable stocks, non-dividend-paying stocks, extreme growth stocks, and distressed stocks. When sentiment is high, on the other hand, these categories of stock earn relatively low subsequent returns.
The efficient market hypothesis gives rise to forecasting tests that mirror those adopted when testing the optimality of a forecast in the context of a given information set. However, there are also important differences arising from the fact that market efficiency tests rely on establishing profitable trading opportunities in real time. Forecasters constantly search for predictable patterns and affect prices when they attempt to exploit trading opportunities. Stable forecasting patterns are therefore unlikely to persist for long periods of time and will self-destruct when discovered by a large number of investors. This gives rise to non-stationarities in the time series of financial returns and complicates both formal tests of market efficiency and the search for successful forecasting approaches.
Financial press reports claim that Internet stock message boards can move markets. We study the effect of more than 1.5 million messages posted on Yahoo! Finance and Raging Bull about the 45 companies in the Dow Jones Industrial Average and the Dow Jones Internet Index. Bullishness is measured using computational linguistics methods. Wall Street Journal news stories are used as controls. We find that stock messages help predict market volatility. Their effect on stock returns is statistically significant but economically small. Consistent with Harris and Raviv (1993), disagreement among the posted messages is associated with increased trading volume
The efficient markets theory reached the height of its dominance in academic circles around the 1970s. Faith in this theory was eroded by a succession of discoveries of anomalies, many in the 1980s, and of evidence of excess volatility of returns. Finance literature in this decade and after suggests a more nuanced view of the value of the efficient markets theory, and, starting in the 1990s, a blossoming of research on behavioral finance. Some important developments since 1990 include feedback theories, models of the interaction of smart money with ordinary investors, and evidence on obstacles to smart money.
Revolutions often spawn counterrevolutions and the efficient market hypothesis in finance is no exception. The intellectual dominance of the efficient-market revolution has more been challenged by economists who stress psychological and behaviorial elements of stock-price determination and by econometricians who argue that stock returns are, to a considerable extent, predictable. This survey examines the attacks on the efficient market hypothesis and the relationship between predictability and efficiency. I conclude that our stock markets are more efficient and less predictable than many recent academic papers would have us believe.
Market efficiency survives the challenge from the literature on long-term return anomalies. Consistent with the market efficiency hypothesis that the anomalies are chance results, apparent overreaction to information is about as common as underreac- tion, and post-event continuation of pre-event abnormal returns is about as frequent as post-event reversal. Most important, consistent with the market efficiency prediction that apparent anomalies can be due to methodology, most long-term return anomalies tend to disappear with reasonable changes in technique. ?
We present a simple overlapping generations model of an asset market in which irrational noise traders with erroneous stochastic beliefs both affect prices and earn higher expected returns. The unpredictability of noise traders' beliefs creates a risk in the price of the asset that deters rational arbitrageurs from aggressively betting against them. As a result, prices can diverge significantly from fundamental values even in the absence of fundamental risk. Moreover, bearing a disproportionate amount of risk that they themselves create enables noise traders to earn a higher expected return than rational investors do. The model sheds light on a number of financial anomalies, including the excess volatility of asset prices, the mean reversion of stock returns, the underpricing of closed-end mutual funds, and the Mehra-Prescott equity premium puzzle.