| October 02, 2013
The authors investigate whether providers of high frequency media analytics affect the stock market.
We investigate whether providers of high frequency media analytics affect the stock market. This question is difficult to answer as the response to news analytics usually cannot be distinguished from the reaction to the news itself. We exploit a unique experiment based on differences in news event classifications between different product releases of a major provider of news analytics for algorithmic traders.
Comparing the market reaction to similar news items depending on whether the news has been released to customers or not, we are able to determine the causal effect of news analytics on stock prices, irrespective of the informational content of the news. We show that coverage in news analytics speeds up the market reaction by both increasing the stock price update and the trading volume in the first few seconds after the news event.
Such coverage also increases prices if the content of the news is positive. Placebo tests and econometric robustness checks, either based on difference-in-difference specifications or different samples, confirm the results. The fact that a provider of media analytics impacts the market in a separate and distinct way from the underlying information content of the news has important normative implications for the regulatory debate.
In recent years, sophisticated traders in financial markets have increasingly used new sources of information such as “sentiment” indicators derived from news wire articles. Such news analytics are computed by computer algorithms and can tell traders within milliseconds whether an article is positive or negative.
In parallel with this, the growth of computerized trading has accelerated the process of accessing such information and increased the speed with which it is incorporated into stock prices. Access to such low-latency information can provide a competitive advantage to its users, which are mainly high frequency and algorithmic traders (e.g., Hendershott and Riordan, 2009, Hasbrouck and Saar, 2013, Baron, Brogaard, Kirilenko, 2012, Brogaard, Hendershott, Riordan, 2012, Boehmer, Fong, and Wu 2012).
However, inaccurate low-latency signals can lead to unintended consequences when algorithms automatically initiate trades based on false information. In April 2013, an incorrect twitter feed about a White House explosion caused a mini flash crash. Some quickly blamed algorithmic trading for the reaction, while others argued that human traders were mainly responsible. In any case, news reading algorithms may be more likely to misinterpret news than human traders.
In July 2013, New York State Attorney General Eric Schneiderman rebuked Thomson Reuters for selling access to key economic survey data two seconds early to high-frequency algorithmic traders. Unlike the early release of such economic survey data, news analytics are based on publicly available news.
Therefore, they constitute a “fairly earned” advantage. However, since news analytics help to trade on public information faster, they give a trader a similar advantage as an early access to private information. In either case, an important question is whether quick-triggered trading initiated by such low-latency information has an impact on the market that is distinct from the underlying informational content of the news. That is, are there potentially distortionary price effects induced by high frequency trading based on news analytics? It seems that only the existence of such distortions should justify regulatory intervention.
This question is very difficult to answer as the response to news analytics normally cannot be distinguished from the reaction to the news itself. We address this identification issue by exploiting a unique experiment provided by differences in product versions of a major provider of news analytics. Because this provider has improved the sophistication of its technology over time, there are differences between older and newer versions.
We use the back-filled analytics of the new version to proxy for the informational content of the news, while traders reacted to the old version that was released at the time. The differences between the old and new version enable us to study the causal impact of news analytics on stock prices. While such differences were rare in a relative sense (roughly 3%), their absolute number of 24,963 is high enough to allow for tests that have sufficient power.
Please use your business email. If you don't have one, please email us at firstname.lastname@example.org.
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 email@example.com. 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.