Parametric versus Non-parametric Bond Pricing and Hedging Models: An Artificial Neural Network Approach to Price and Hedge US Treasury Securities

Financial institutions assets are quite sensitive to fluctuations in interest rates. The resulting interest rate risk can be managed with a model that accurately prices and hedges interest rate exposure for a portfolio of US Treasury zero-coupon bonds. This book compares the parametric and non- parametric (using artificial neural networks) approaches for pricing and hedging a US Treasury zero-coupon bond portfolio. The parametric pricing models considered are the Cox Ingersoll and Ross (1985a,b) model, Longstaff and Schwartz (1992) model, and restricted Heath Jarrow and Morton (1992) models (Ritchken and Sankarasubramanian (1995)). The neural network models include the multi-layer perceptron and the radial basis function networks. The comparison is done on two grounds, their ability to price zero-coupon bonds and to hedge the underlying risk factors. Risk managers, fixed income traders and academicians can refer to this book for detailed mathematical analysis and in-depth coverage of different interest rate models, their estimation and application to price and hedge interest rate risk.