The 2019 Plato Partnership MI3 academic conference was held at the Imperial College London Business School in conjunction with Imperial College and the Centre for Economic and Policy Research (CEPR). One of the keynotes was delivered by BMLL Technologies about feature generation at scale.
Specializing in the limit order book data set, BMLL hosts historical data for all global liquid electronically traded securities. The limit order book data is processed to a common format and held on a cloud platform.
Open source has enabled significantly increased adoption over the last 15 years, with early aggregation of data taking place on Microsoft Excel, processing only 300 aggregates. This has increased complete analysis of the orderbook as of 2018, with cloud hosted, scalable solutions integrating machine learning dominating the world of data.
A new algo-driven paradigm has resulted in data science and machine learning being placed at the core of every market participant’s practice. Raw trade data is normalized, analyzed, input into machine learning functionality and produces research.