Explainable AI in Request-for-Quote

Explainable AI improves prediction accuracy for financial markets: The paper demonstrates that advanced machine learning models (including Random Forest, XGBoost, and Bayesian Neural Trees) can more accurately predict whether financial quotes will be accepted, helping market makers set better prices and improve overall market efficiency.

Transparency matters in high-stakes financial decisions: By using explainable AI rather than "black box" models, the research shows that market participants can understand why the system makes specific predictions about quote fulfillment, enabling more informed decision-making in less liquid asset classes where stakes are particularly high.

XAI addresses a real market problem: The paper identifies that accurately predicting Request-for-Quote (RFQ) fill rates is "crucial for improving market efficiency" in less liquid markets, suggesting that explainability helps solve a genuine challenge that affects financial market participants and potentially broader economic efficiency.

Multiple AI approaches work together: Rather than relying on a single algorithm, the research combines several machine learning techniques (Logistic Regression, Random Forest, XGBoost, and Bayesian Neural Trees), suggesting that responsible AI implementation may benefit from ensemble approaches that leverage different model strengths.