| May 02, 2019
In a recent report, Deutsche Bank took a novel approach to using RavenPack news analytics data for estimating equity positioning based on the time-varying reaction to news flow.
The key takeaway from the report is that a better understanding of the marginal investor sentiment and positioning can help improve longer-term trading signals, especially when it comes to price momentum and earnings surprise.
The report aims to decompose news sentiment to detect crowded long and short stocks by evaluating relative price moves against positive and negative events. They examine a pure sentiment-based strategy which produced a low-turnover long-short annualized return of 7% in Australia and Europe, 5% in Asia ex Japan, and 3% in the US. Perhaps more interestingly, they examine an augmented momentum strategy using this sentiment signal, as well as incorporating a crowdedness metric into earnings-based strategies.
While much of the substance that the report seeks to test seems intuitive, working through hundreds of thousands of news articles that hit the tape every day is untenable for a fundamental analyst. The approach DB took is interesting, in part, for its effort to estimate the more subtle effects of positioning and crowding to be incorporated into an investment strategy, rather than trading on the sentiment signal alone.
The following illustrative charts from the report help demonstrate the main goal:
General intuition holds that price performance, in excess of factor returns, would be symmetric around the origin (left chart). However, observed price action shows that this is not always the case. We observe that, at times, certain assets don’t react to news the way we would expect them to. Namely, the reaction to good news is often outsized relative to reactions to bad news and vice versa.
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.