RBC Capital Markets has launched an AI-powered electronic trading platform it claims will help improve results and insights for clients in a "measurable and explainable way".
Built with the Canadian bank's Borealis AI research centre, the platform, called Aiden, looks to apply deep reinforcement learning into a constantly changing environment like equities trading.
RBC says Aiden has shown the ability to navigate the challenges of fluid and dynamic market conditions in real-time, without the need for continuous re-coding like traditional trading algorithms.
It taps hundreds of pre-programmed data inputs and its ability to make more than 32 million calculations per order to execute trading decisions based on live market data, dynamically adjust to new information and learn from each of its previous actions.
Shary Mudassir, co-head, global equities execution, RBC Capital Markets, says: "The market volatility we have seen during Covid-19 has been a great proving ground for Aiden. Aiden was able to recognise the sudden changes in the market and adapt to them, allowing it to preserve performance."
Its first live offering, a volume-weighted average price algorithm, was designed to help improve clients’ trading performance by reducing slippage - the difference between the expected price of a trade and the price at which the trade is executed.
It is already available to RBC Capital Markets clients who are trading in US and Canadian equities, with plans to introduce the platform to additional clients across geographical markets in the next few months.