Technology solutions provider CQG has completed internal testing and proof-of-concept for its new AI predictive model toolkit, designed to predict futures market moves.
With the product, which the company believes to be the first of its kind, CQG aims to improve retail traders’ and buy-side firms’ ability to identify trading and analytics opportunities, guide trading strategies and manage their positions.
The firm reports “an extremely high level of predictive success” in the solution’s anticipation of futures market moves, with 80% accuracy on whether the next move in the E-mini S&P 500 futures contract would be up, down or unchanged. This matches levels seen in the back-testing environment, the company says.
Kevin Darby, vice president of execution technologies, outlines the challenges faced by the company during development, including “storing and curating terabytes of historical market data while retaining the ability to make decisions in microseconds in real-time environments.”
He continues: “We built bridges between the current machine learning (ML) infrastructure, based on the Python language, and the reliance of the financial industry infrastructure on C++. We also needed to recast the traditional ML training pipeline to optimise for generative time series prediction to estimate conditional probability distributions in a mathematically satisfying and stable way.”
Ryan Moroney, CEO of CQG, explains that clients will be able to use the firm’s ML lab and cloud computing resources to either build on CQG’s foundational models or build their own, using the company’s ML toolkit and historical data.
The firm states that a number of algorithm, charting and research-related applications for the new solution have been established, adding that further use cases are being explored with partners.
©Markets Media Europe 2024
