Applying Heath-Jarrow-Morton Model to Forecasting the US Treasury Daily Yield Curve Rates
Applying Heath-Jarrow-Morton Model to Forecasting the US Treasury Daily Yield Curve Rates
复制标题
应用 Heath-Jarrow-Morton 模型预测美国国债每日收益率曲线利率
作者：
V. Maltsev;Michael Pokojovy
Heath-Jarrow-Morton (HJM) 模型是描述无套利假设下利率曲线随机演化的强大工具。 HJM 方法的一个重要特征是漂移可以表示为各自波动率和潜在相关结构的函数。本文面向研究人员和实践者，旨在对建立在兴趣理论和无限维随机偏微分方程基础上的抽象 HJM 框架进行独立而简洁的回顾。为了说明该理论的预测能力，我们将其应用于美国国债每日收益率曲线利率的建模和预测。我们将非参数模型与美国财政部提供的实际数据进行拟合，并说明其在预测未来收益率曲线利率方面的统计表现。
The Heath-Jarrow-Morton (HJM) model is a powerful instrument for describing the stochastic evolution of interest rate curves under no-arbitrage assumption. An important feature of the HJM approach is the fact that the drifts can be expressed as functions of respective volatilities and the underlying correlation structure. Aimed at researchers and practitioners, the purpose of this article is to present a self-contained, but concise review of the abstract HJM framework founded upon the theory of interest and stochastic partial differential equations in infinite dimensions. To illustrate the predictive power of this theory, we apply it to modeling and forecasting the US Treasury daily yield curve rates. We fit a non-parametric model to real data available from the US Department of the Treasury and illustrate its statistical performance in forecasting future yield curve rates.