Kay Giesecke, CEO and Founder of Infima Technologies, spoke at the AI & Data Science in Trading Event on September 8, 2021. Infima’s transformative Deep Learning technologies deliver prepayment predictions at borrower, security, and cohort levels that set new accuracy standards in the Agency MBS market, empowering investors to make superior security selection and portfolio construction decisions. Innovative analytical tools support a range of workflows, from powerful querying to in-depth evaluation of securities.
The presentation entitled "Deep Learning for MBS Prepayments" was based on the material discussed in this Infima White Paper. Here's the abstract of the presentation:
Predictions of prepayment speeds are mission-critical for investors, dealers, originators and other participants in the $10T Agency MBS market. Legacy prediction technologies produce flawed and stale data wrecking returns and profits. We develop deep learning systems that set new accuracy and latency standards, delivering performance-boosting edges to market participants. Our systems harness data of unprecedented size and granularity, covering monthly records for tens of millions of borrowers across the US over two decades. By uncovering hidden nonlinear patterns in borrower behavior at the individual loan level, they improve prediction accuracy for MBS pool CPRs by a full order of magnitude relative to the market’s current “gold standard.” Our predictions are robust in all market environments including the pandemic. Rigorous significance tests offer deep insights into the variables influencing predictions.
Today, the combination of endless new data sources, cheap computing and new AI techniques is powering fundamental change - it’s time to ‘raise the bar’ in terms of technology. AI & Data Science in Trading is the event for senior management from hedge funds and investment banks to discover how to maximize this opportunity. This is the largest focused meeting of experts within Capital Markets using AI for alpha discovery, managing risk and optimising portfolios, and takes place annually in both New York and London and now online. Twitter: @aidataconf