Big Data Equities
RavenPack | November 17, 2020
In this study, RavenPack data scientists sought to isolate and quantify the additional value provided by news sentiment both by itself and on top of other more traditional factors in trading Asia Pacific equities.
Whether news sentiment - aggregated from online news stories and social media posts - could enhance a traditional multi-factor Asia Pacific (APAC) equities portfolio quantitative investment strategy (QIS), was the subject of this White Paper.
Researchers began by constructing a News Sentiment Index to analyze the effectiveness of sentiment on its own; in step 2 they would seek to test how well it performed as an overlay to a traditional factor QIS.
The Index used sentiment measured over a variety of smoothing periods, ranging from 1 week to 90 days (3 months) to analyze the impact of both short-term and longer-term sentiment.
Differing portfolio rebalancing periods were also experimented with from daily to monthly.
The Sentiment Index’s signals were then applied to a long-short, market, and factor neutralized APAC equities portfolio.
The results were overall promising. The index smoothed over a 90-day period, with monthly rebalancing, delivered an IR above 0.7 and an Annualized Return above 400bps for an AUM of $1B, with an Annualized Turnover under 1500%.
Next, researchers wanted to see whether the signals derived from sentiment could enhance a multi-factor QIS and if so by how much?
The traditional multi-factor framework for Asian markets consisted of Momentum, Low Size, Low Volatility, Growth, Value, and Quality factors.
An analysis using MSCI’s GEM3 model factor exposures framework suggested sentiment was fairly uncorrelated to other factors, “especially over the short aggregation windows, which correspond to the fastest signals.” According to Peter Hafez, Chief Data Scientist at RavenPack. This held out the promise sentiment could add value.
With the addition of sentiment, multi-factor models showed consistent improvement. As an overlay to a $1B AUM strategy, sentiment increased the IR by 64%, from 0.45 to 0.74, and the Annualized Return by 35%, from 167bps to 225bps. A further advantage of the multi-factor approach was the lighter turnover and slightly larger portfolio sizes.
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.
Schedule a personalized trial to see how our actionable insights can boost your investment strategies.
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.