.jpg)
BMLL, the leading, independent provider of harmonised, historical data and analytics, today announced a collaboration with New York University’s Mathematics in Finance program, making its Data Lab available to its team of quantitative researchers.
Part of NYU’s Courant Institute of Mathematical Sciences, the program and its research activities are directed by Professor Petter Kolm, a leading quantitative analyst specialising in market microstructure modelling and buy-side trading, who was named as “Quant of the Year” in 2021 by Portfolio Management Research (PMR) and Journal of Portfolio Management (JPM) for his contributions to the field of quantitative portfolio theory. An NYU team, led by Professor Kolm, has previously conducted research that showed that deep-learning models across a large number of stocks at large scale is found to be fully feasible, and that future returns of “information-rich” stocks can be predicted more accurately by deep learning.
The BMLL Data Lab is a data science platform that allows users to access over 3 years of Level-3 harmonised futures data, process it at scale, and find inferences by drilling down into every single message.
Professor Petter Kolm, NYU Courant Professor, said “through the BMLL Data Lab, we can access pristine Level 3 Data and run computation at scale to work through huge data sets in a fully managed and documented environment”.
Paul Humphrey, CEO of BMLL, said: “We are delighted to collaborate with this NYU program and support Professor Petter Kolm and his colleagues in carrying out their cutting-edge limit order book research, and further developing the research they conducted on equity markets into futures exchanges. This collaboration is an acknowledgement of the quality and depth of our Level 3 order book data, and how our Data Science as a Service solution makes it possible to rapidly achieve actionable results.”
Dr. Elliot Banks, Chief Product Officer, BMLL, said “BMLL combines easy to use APIs and analytics libraries in a secure cloud environment, allowing scalable research without the burdens of data sourcing, curation or engineering. We are proud that our tool has been selected by one of the world’s best researchers to further explore the predictive power of historic data.
