ING launches AI tool Katana Lens to help bond investors make sharper decisions
Co-created with Dutch pension fund PGGM, Katana Lens is a web-based application that leverages predictive analytics to help bond investors make faster and sharper decisions within minutes.
New solutions based on artificial intelligence (AI) continue to gain adoption in the financial services sector, with another piece of proof for that trend just provided by ING. The company has announced the launch of Katana Lens, an AI tool which aims help investors to easily find and compare trade ideas.
Co-created with Dutch pension fund PGGM, Katana Lens is a web-based application that harnesses the capabilities of predictive analytics to help bond investors make faster and sharper decisions within minutes.
“What used to take forever, now only takes five minutes and a cup of coffee,” said Santiago Braje, global head of Credit Trading, who announced Katana Lens today at the Artificial Intelligence Summit.
Katana Lens uses an algorithm that learns from the history of hundreds of thousands of trades and identifies the most promising ones. By going through all the pairs of bonds and taking each ‘buy’ and ‘sell’ combination as a possible investment, it simplifies the selection process for investors, who are presented with a prediction or suggested decision.
The launch of Katana Lens follows up on the launch of Katana in December 2017. Katana was created to help traders decide what price to quote when buying and selling bonds for their clients based on historic and real-time data. While Katana is a tool to improve pricing as well as the speed of the traders’ response on the sell-side (banks) within the bond market, Katana Lens was developed for the buy-side (investors).
The first results of testing Katana with the emerging markets desk in London show faster pricing decisions for 90% of trades; reduction in trading cost by 25%; and traders are able to offer clients the best price four times more frequently.
The launch underlines the advances of AI in the financial services sector. In May this year, Japanese fintech expert AlpacaJapanCo., Ltd. announced a partnership with Bloomberg L.P. and the launch of a price forecasting application “AlpacaForecast AI Prediction Matrix” for Bloomberg users. In particular, Alpaca applied convolutional neural network (CNN) based deep learning engine with high pattern recognition function commonly used for image recognition/video recognition to time series analysis. Analysis of tick data which has been hard for human eyes to recognize makes market prediction possible. The list of initially supported assets includes USD/JPY, EUR/USD, AUD/JPY and US Treasury 10 years bond etc.