Arbitrage-Free Bond and Yield Curve Forecasting with Neural Filters under HJM Constraints
This addresses the problem of ensuring arbitrage-free predictions in financial markets for practitioners, though it is incremental as it combines existing methods like neural networks with established HJM models.
The paper tackled forecasting bond prices and yield curves without arbitrage by embedding HJM constraints into a neural state-space model with arbitrage error regularization, showing improved accuracy at short maturities, such as increased bid-ask hit rates and reduced prediction errors in 5-day forecasts.
We develop an arbitrage-free deep learning framework for yield curve and bond price forecasting based on the Heath-Jarrow-Morton (HJM) term-structure model and a dynamic Nelson-Siegel parameterization of forward rates. Our approach embeds a no-arbitrage drift restriction into a neural state-space architecture by combining Kalman, extended Kalman, and particle filters with recurrent neural networks (LSTM/CLSTM), and introduces an explicit arbitrage error regularization (AER) term during training. The model is applied to U.S. Treasury and corporate bond data, and its performance is evaluated for both yield-space and price-space predictions at 1-day and 5-day horizons. Empirically, arbitrage regularization leads to its strongest improvements at short maturities, particularly in 5-day-ahead forecasts, increasing market-consistency as measured by bid-ask hit rates and reducing dollar-denominated prediction errors.