| October 27, 2017
Recently, RavenPack hosted its 5th Annual Research Symposium in New York titled “The Big Data & Machine Learning Revolution”. Here are my personal take-aways.
Following the success of last year’s event, this year we had over 700 people register to participate in our annual research symposium. We had to turn many of those away, since our capacity in the room was limited to 250 attendees.
presentation slides and video recordings
from the event.
First, I suggest you watch this 3 minutes event highlights video for a brief overview of the presentations.
The event started off with our own
Armando Gonzalez, RavenPack CEO
, going through the concepts of Big Data and Machine Learning in finance, and arguing why we are experiencing a
Big Data and Machine Learning Revolution. Armando highlighted four areas that are being disrupted by artificial intelligence (AI) and machine learning, including access to unstructured content, the data modeling/cleaning process, the generation of alpha signals (predictive analytics), as well as portfolio optimization.
Following Armando’s opening address, we continued with our keynote presentation by
Marko Kolanovic and Rajesh T. Krishnamachari, the Global Head and Vice President of the Quantitative and Derivatives Strategy at J.P. Morgan
, respectively. They provided an overview of their highly celebrated tour de force titled
“Big Data and AI Strategies”, in which they put a host of alternative data providers, including RavenPack, under the microscope. They also presented their team’s latest research which used
sentiment data from RavenPack to enhance valuation strategies.
Miquel Noguer Alonso, Executive Director, UBS and Adjunct Assistant Professor at Columbia University
then went on to explore the
benefits and challenges of different modeling choices in the signal construction process,
covering anything from fast machines, to supervised, unsupervised, and deep learning models. He addressed the challenge which arises when discerning how to use the various predictions to create the final portfolio weights. You can learn more about this aspect from Miquel’s presentation.
Next, on behalf of Ravenpack, I had a chance to present some the internal research done by the RavenPack Data Science team, looking at
strategies around energy and metals futures using machine learning and event detection.
In particular, I presented how, using ensemble methods, we were able to achieve attractive Information Ratios across both commodity baskets with holding periods of several days.
After this, we then had the first exciting panel of the day, focusing on how to prepare your organization for the big data revolution. The moderator,
Adam Honore from the CME Group
, started off the session on a high by challenging the audience to get the panelists “off their game”. The audience was quick to respond with many thought-provoking and challenging questions, although they didn’t succeed. Instead,
Thani Sokka from Dow Jones, Stuart Kozola from Mathworks, and our own RavenPack CTO, Jason Cornez
had a lively discussion about how the “cloud” has revolutionized their own businesses and how it will revolutionize finance. The question that caught my attention the most was one asking if the panelists had seen any “resistance” to this revolution.
Ichihan Tai, Head of Data Science at Tokio Marine Asset Management
spoke about the concept of
lean research, using news analytics
to replace more expensive human-driven processes related to information gathering, thereby reducing overall R&D costs. In particular, he walked through one example, using RavenPack data to identify boring vs. popular stocks (“the boring stocks premium”) and showed how this approach provided alpha beyond analyst data from IBES as well as traditional factors. That was the presentation I personally liked the most.
Hedi Benamar, a Senior Economist for the Board of Governors of the Federal Reserve System
, who talked about the greater impact of macroeconomic news on treasury bond prices when demand for information around these events are high.
James Hodson, from the Artificial Intelligence Laboratory at the Jozef Stefan Institute
, discussed his recent work leveraging
large-scale employment data to better understand firm performance, evolution, and labor markets.
He found that firms with lower employee turnover systematically outperform those with higher turnover rates.
Jason Cornez, RavenPack CTO
, went on to introduce RavenPack’s new custom indicators platform
that allows users to easily apply their own filters and aggregators on top of the granular RavenPack data, thereby consuming a single daily snapshot of sentiment or media buzz, delivered via API or through the RavenPack Visual Platform.
Donald Putnam, Managing Partner, Grail Partners
gave his opinions on artificial intelligence, focusing on where we have been, and where we might be going. Rather than focussing on the immediate threats and opportunities, Donald related long-standing themes and extended them in a seven to ten-year view. Take a look at his
very engaging presentation.
We finished off an exciting day with a highly entertaining panel on
alternative data and why we should care, moderated by
Marko Kolanovic from J.P. Morgan
. Panelists included
Matthew Rothman, Managing Director and Head of Quantitative Equity Research at Credit Suisse, Yin Luo, Vice Chairman at Wolfe Research, Donald Putnam, Managing Partner at Grail Partners
, and myself. We all shared our opinions on datasets that we had found either particularly useful or disappointing as part of our own research. In particular, Matthew Rothman shared a very entertaining analysis using Google searches to predict unemployment levels, which produced rather dubious, but amusing correlations. The final comment of the panel concluded with: “I’d rather have a drink than an opinion”.
Please use your business email. If you don't have one, please email us at email@example.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 firstname.lastname@example.org. 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.