| July 20, 2009
The purpose of this paper is to investigate the impact of news sentiment on abnormal stock return, which allows us to focus on company specific news events as opposed to news driving the overall market
The analysis is conducted on a portfolio of stocks including the constituents of the Dow Jones Industrial Average and the Eurostoxx50 covering the years 2005 through 2008. Based on a set of sentiment classifiers used to process textual input of news stories, we construct a set of events characterized by unique combinations of sentiment classifications.
The degree of rationality in financial markets has long been a topic for discussion in both academic and professional circles. Rationality implies that investors correctly use all available information in establishing security prices, and hence a natural consequence of this definition is that at least unexpected news can be expected to impact the information set and thus also the securities pricing mechanism. Taking into consideration the version of the efficient market hypothesis (EMH), where the costs of producing and processing information are explicitly recognized, the price of a security at any point is simply a noisy estimate of the present value of the certainty equivalents of its risky future cash flow.
An assumption underlying the EMH is that investors learn to make correct inferences about the impact of new information on the probability distribution of potential stock returns - that is, they form rational expectations about the future . According to K.C. Brown et al., the usual definition of rationality does not imply that securities prices react to major informational surprises instantaneously. For instance, even when an event clearly conveys good or bad news about a firm’s or the market’s prospect, the full extent of its eventual impact on stock prices may be uncertain.
Thus, with incomplete information, the best that investors may be able to do is to estimate the parameters of a conditional probability distribution summarizing the various potential outcomes. With the introduction of the uncertain information hypothesis (UIH), they argue that in the aftermath of new information, both risk and expected return of the affected companies change in a systematic fashion.
Many economic models assume that people are at least on average rational, and can in large enough quantities be approximated to act according to their preferences. In contrast, the concept of bounded rationality revises this assumption to account for the fact that perfectly rational decisions are often not feasible in practice due to the finite computational resources available for making them. Simon suggests that economic agents employ the use of heuristics to make decisions rather than a strict rigid rule of optimization. They do this because of the complexity of the situation, and their inability to process and compute the expected utility of every alternative action. Deliberation costs might be high and there are often other, concurrent economic activities also requiring decisions .
Introducing the concept of bounded rationality and general behavioral finance theory opens up for the possibility that human decisions may be influenced by other factors than economic and fundamental information. Instead, market sentiment may be considered a vital part in the securities pricing mechanism. More specifically market sentiment, often called the ”market consensus”, is the favorable or unfavorable mood that prevails at a given moment among investors and analysts about the future market price evolution for instance of specific companies or the market as a whole. Market sentiment, as such, might be acquired from more than one sentiment analytical tool. For example, the tools include simple extractions of movements of stock or futures prices or extractions of news and media information based on their polarity. Alternatively, it may also be possible to extract information about sentiment from the options pricing mechanism. For sure there are many ways on how to extract the sentiment or trend of the market. More recently, investors are known to measure market sentiment through the use of news analytics, which include sentiment analysis on textual stories about companies and sectors .
In this paper, we will stay away from making direct comments on the rationality of financial markets, instead we will focus on investigating how market sentiment, as reflected in this paper solely by news sentiment, may impact the abnormal stock return in the minutes following company specific unscheduled news announcements. As part of this analysis, we condition on the set of sentiment classifiers provided by RavenPack, as well as on a relevance factor measuring the relevance of a particular news story in relation to a specific company.
In Section 2, we present the definition of abnormal returns as given by the Capital Asset Pricing Model. In Section 3, we present our definition of an event category given a set of sentiment scores. In Section 4, we describe our data. In Section 5, we present our empirical results, and finally in Section 6, we present our conclusion and suggestions for further research and refinements.
The risk of a portfolio or single equity comprises systematic and unsystematic risk. Systematic risk refers to the risk common to all securities i.e. market risk and hence cannot be reduced through diversification, while unsystematic risk is the risk associated with individual assets. Since company-specific news announcements are expected to impact at most a subset of equities, such announcements are to be considered part of the unsystematic risk component. Borrowing from the event study technique, it is possible to isolate the effect of the stock specific reaction to a news event by focusing on abnormal rather than realized returns. Several techniques exist on how to measure and adjust for what is known...
Please use your business email. If you don't have one, please email us at email@example.com.
We will process your personal data with the purpose of managing your personal account on
RavenPack and offering our services. You can exercise your rights of access, rectification,
erasure, restriction of processing, data portability and objection by emailing us at firstname.lastname@example.org. For more information, you can
Your request has been recorded and a team member will be in touch soon.
High inflation has returned in developed markets after decades of lying low. In our latest paper, we show how to build an inflation-based asset allocation strategy using sentiment data and we illustrate that sentiment-based strategies outperform models that depend merely on past observed inflation values.
This year's RavenPack Research Symposium brought two intense days of knowledge sharing in London and New York, from 25 top experts in natural language processing, quantitative investing and machine learning. Together, we explored how firms can leverage new language models to generate alpha, better manage risk and respond to calls for more sustainable investment practices.
Human capital is at the heart of value creation. Our latest research demonstrates how unprecedented workforce insights, sourced from over 200 million job postings, can generate more alpha.