Bloomberg brings early alerts to corporate credit with machine learning

Early alerts will provide traders with predictive insights into USD corporate investment grade and high-yield securities.

Bloomberg has rolled out early alerts using machine learning technology to provide clients with predictive insights for fixed income trading.  

The data-driven early alerts use Bloomberg’s proprietary library of fixed income data alongside machine learning models for predictive intelligence that could offer users an edge in trading corporate credit markets.

By analysing more than 16,000 US dollar-denominated corporate investment grade and high-yield securities, the early alerts will provide scores over one, five and 20-day periods, and offer insight into the likelihood a corporate security will see significant credit spread tightening or widening over the specific time periods.

“Fixed income markets, including corporate credit, are primed for the type of quantitative modelling, tooling and predictive power already available in other asset classes,” said Brad Foster, global head of enterprise content.

“Early Alerts, offering probable indication of whether a corporate bond’s spread will widen or tighten, provides traders, asset managers, portfolio managers and risk managers a signal that easily compliments and enhances their existing process and a dataset that they can easily integrate into their models to more accurately anticipate market movement, further enabling them to meaningfully inform their investment strategy.”

The early alerts, which will be accessible to Bloomberg Enterprise Data clients, will initially cover the US dollar-denominated corporate issuers but will expand into other regions that cover issuance in other currencies in the near future.

“Our rich ongoing and historical datasets, robust back-testing process, machine learning and subject matter expertise help to collectively produce a signal that will enable our clients to add cutting edge predictive power to an asset class that has, to this point, lacked investment and access to such analytics,” Foster added.