By Lee Bray, Asia Pacific Head of Equity Trading, J.P. Morgan Asset Management
The firm’s Asia Pacific equity trading team aims to have around 50% of trading automated with machine learning by the end of this year.
J.P. Morgan Asset Management has invested significant resources in building machine learning tools to enhance global equity trading, as the rise of artificial intelligence and automation transforms how we conduct business.
The firm’s Asia Pacific buy-side equity trading team developed a model utilizing machine learning to help its portfolio managers make the execution of trading orders more effective and cost efficient. The proprietary model, which was developed by the firm’s quantitative analysts and traders, uses data patterns to find the optimal execution strategy for trading orders. The Asia Pacific trading desk is on track to have 50% of equity trading flow driven by machine learning by end of 2018
With myriad options available for executing any given order, particularly smaller or more routine orders, an intelligent model can identify the best execution more efficiently than a human. The artificial intelligence model “learns” constantly about the best outcomes for trading orders and adapts by recalibrating as market conditions change and new information is delivered. Using these algorithms to pinpoint the probability of best performance, the machine learning environment auto-routes and executes accordingly.
To develop the model with machine learning, we tapped into techniques more commonly found at companies like Facebook and Google. By creating a systematic, adaptive model able to alter actions based on mathematical patterns rather than relying on human input, we’re transitioning equity trading to be more scientific and quantifiable.
Currently the model gives trading recommendations to human traders, but increasingly it is taking over an automated role in executing trades. As its application becomes more scalable, it has the potential to generate cost savings for JP Morgan Asset Management’s clients through greater trading efficiency.
We’d love to hear your feedback on this article. Please click here