Equities
RavenPack | February 26, 2013
As part of this paper, we move beyond the simple notion of "buy on positive and sell on negative sentiment"
Instead try to understand the interaction effects between sentiment and various forms of media or investor attention. To understand how investors react to sentiment, we conduct a double sort analysis - ranking on sentiment in one dimension and on media coverage, sentiment momentum, return volatility, and price momentum in the other.
We find that conditioning news sentiment impact with attention yields higher returns.
Specifically:
The emergence of news analytics has been valuable to both academics and financial practitioners in understanding the impact of news and public information on asset pricing. By now, research has led to a set of stylized facts that go beyond describing news flow characteristics. For instance, Chan [Mar, 2001] found that negative news leads to stronger price reactions than positive news - a result further supported by Cahan et.al. [May, 2009]. In the paper "Stock Price Reaction to News and No-News: Drift and Reversal after Headlines" [Mar, 2009], Chan found that negative news tends to be reflected more slowly in stock prices than positive news. In Hafez and Xie [May, 2012], we found that positive and negative events on average decay over a two and five day horizon, respectively. Several studies have also shown that price reactions to news are not only stronger for small cap than for large cap stocks, but that news based signals also decay more slowly as one moves down the size scale.
Even though the body of research continues to grow, there is still a lot to learn about how investors react to news and sentiment over time. Do investors react differently during bull or bear markets? Or when investing on different sectors or according to various company characteristics? The objective of this paper is to move beyond the simple notion of “buy on positive and sell on negative sentiment” and instead try to understand the interaction effects between sentiment and various forms of media or investor attention. To understand how investors react to sentiment, we conduct a double sort analysis - ranking on sentiment in one dimension and on various forms of attention in the other including media coverage, sentiment momentum, return volatility, and price momentum.
In the following section, we provide a brief description of our test stock universe and the data used in the study. In Section 3, we present a new methodology to calculate the company sentiment indicator tested here. In Section 4, we consider the value add of including sentiment decay as part of our indicator methodology. In Section 5, we look at the interactions effects between sentiment and attention; and evaluate its effects on various industries in Section 6. Finally, in Section 7 we present our conclusions.
Please use your business email. If you don't have one, please email us at info@ravenpack.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 privacy@ravenpack.com. For more information, you can check out our Privacy Policy.
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.