November 8, 2022
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.
If you were unable to attend, here's a summary of the key takeaways, with links to some of the speakers' presentations.
CEO & Co-Founder
Language AI is poised to become the next disruptive innovation, resembling in their own times, those of the telegraph, the telephone or the personal computer.
In his keynote presentation, Language AI - The Age of Disruption, Armando Gonzalez, Co-founder and CEO of RavenPack argued that the recent developments in natural language processing are on track to fundamentally change how we work:
We're at the tipping point of disrupting how we extract deeper knowledge from data to make more informed decisions
The context has never been more exciting: powerful computing, cheap storage, and vast amounts of data which, to a large extent (>80%) is unstructured.
Huge advancements in Machine Learning and Natural Language Processing have made it possible to feed hyper focused training sets into machines, thus enabling them to help solve specific problems across the finance, business and public sectors.
Global Head of Quantitative R&D
Abu Dhabi Investment Authority (ADIA)
Investing can be characterized as a data science problem. The case for building a lab and factory for scientific investing
Investment firms hire specialists, but entice them to become generalists (i.e., portfolio managers). This is preventing quantitative strategists from achieving their full potential. In a fireside chat with Armando Gonzalez, Marcos Lopez de Prado, Global Head of Quantitative R&D at Abu Dhabi Investment Authority (ADIA) argued that, although investment firms have attracted scientific talent, they have done a poor job at developing it. A research lab structure would offer a unique environment for developing scientists, allowing them to tackle well-defined open investment problems.
Chief Data Scientist
Language AI drives alpha by providing additional context and reducing noise in investment strategies across multiple holding periods and asset classes.
Language AI has a wide range of applications beyond simply gauging sentiment. The most powerful trading indicators account for context, and help investors better understand the half-life of a sentiment signal. In his talk, Thematic Boosting of Investment Strategies Using Language AI, Peter Hafez, RavenPack's Chief Data Scientist showed how to leverage the RavenPack Event Taxonomy to create three thematic signals including Earnings Intelligence, ESG Controversy, and Economic Activity. Get his deck here.
Chief Strategy Officer
Jobs intelligence is a highly underrated source of insights for the investment world
Keeping an eye on hiring trends, shifts in the most sought-after skills, or on hiring locations can give investors valuable insights into performance and investment prospects. Aakarsh Ramchandani, RavenPack's Chief Strategy Officer, introduced our newly released Job Analytics dataset, sourced from over 200 million job postings, in partnership with LinkUp. He also gave a snapshot of the latest research paper leveraging this data, which proves, among many other things, that monthly hiring growth is indeed positively correlated with future stock performance. See his deck here: Drawing Value From Job Data.
Professor of Finance
Georgia Southern University
To deal with unscheduled events that hit stocks, investors engage in narratives which simplify the complexity of real-time change.
Nicholas Mangee, Professor of Finance at Georgia Southern University described “animal spirits'' - instincts and emotions that drive human economic behavior - and its application in the emerging field of narrative economics. Based on his recent book, Pr Mangee described the Novelty- Narrative Hypothesis for the U.S. equity market by conducting a comprehensive investigation of unscheduled events using big data textual analysis of financial news. See his presentation here.
Ultimately, it's about having a fundamental thesis and linking it with the power of big data and machine learning
…argued Bogdan Ianev, Managing Director at Credit Suisse, during a panel on Applications of Machine Learning, quoting a 2018 paper co-signed by Nobel laureate Harry Markovitz. And as an example of how traditional thinking can go along with NLP and AI, Bogdan described the success of the Credit Suisse RavenPack AI Sentiment Index, which is powered by RavenPack's news sentiment. The index is based on a simple concept - create a sector rotation strategy across US large caps that will rely on news sentiment data, which has been proven to have predictive power over long periods of time. This is also backed by the Index's performance so far.
Fun fact: the processing power of today’s iPhone is comparable to that of a super computer in 2001, while data storage is now close to a penny a gigabyte versus a penny a megabyte 20 years ago.
Senior Data Scientist
Language AI helps predict real time economic activity, including GDP and inflation rate
News and sentiment analytics can significantly improve timeliness and accuracy of traditional macroeconomic nowcasting models. Paolo Andreini, Senior Data Scientist at RavenPack introduced a macroeconomic nowcasting model based on RavenPack EDGE to predict real-time economic activity. Using the RavenPack Event Taxonomy, RavenPack’s data science team built granular indicators that captured the most relevant economic features. As research showed, this delivered higher returns across alternative passive strategies. See his detailed arguments in this deck.
ESG Quantitative Researcher
The road from ESG intent to ESG achievements remains murky. News analytics help investors better analyze, monitor, and qualify the sustainability profiles of companies
Speakers from Dow Jones, BlackRock and Capital Fund Management highlighted the growing role of data-driven processes to achieve scale and efficiency across ESG applications. RavenPack's ESG Quantitative Researcher, Ludovic Mathieu, explained how our ESG Controversy Scoring framework detects real time events across more than 40,000 news sources in 13 different languages. To illustrate the framework, RavenPack’s PowerBI experts created a visualization suite that enables investors to track their portfolio exposure to sustainability-linked controversies.
Financial Services Industry Specialist
Amazon Web Services
Financial institutions are increasingly investing in AI and Machine Learning because it is cost-effective, easy-to-use, and scalable
Mark O'Donnell, Financial Services Industry Specialist at Amazon Web Services described how technology is transforming the financial services industry, with applications in customer service, document processing, credit decisioning and underwriting, fraud prevention, predictive analysis and personalized recommendations. See his deck here.
AWS Principal Solution Architect
The evolution towards cloud native risk computing
Richard Nicholson, AWS Principal Solution Architect FSI BD introduced the AWS HTC initiative – a configurable cloud native blueprint platform enabling high throughput / high scale out market risk calculation. He explained the design tenets, the implementation and the relevance with reference to the next generation of algorithmic techniques. See his presentation here.
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.