基于远期利率分解技术的三因子HJM模型研究  

Empirical study of three-factor HJM model based on forward rate decomposition technique

作　　者：李彪[1] 杨宝臣[1] 

机构地区：[1]天津大学管理学院,天津300072

出　　处：《管理科学学报》2008年第6期112-121,共10页Journal of Management Sciences in China

基　　金：国家自然科学基金资助项目(70471051)

摘　　要：在 HJM 框架下远期利率期限结构可以分解成两个成分函数,其中一个表示观测到的初始远期利率曲线,另一个表示远期利率的动力学演变过程.由于两者具有相同的参数,因而采用这种分解技术可以简化 HJM 类模型中参数的估计过程.为有效拟合远期利率曲线的形状,在原两因子 HJM 模型的基础上又引入了另外一个价差因子,并利用该三因子 HJM 模型的泛函性,推导得到一个简单有效的参数估计程序.据此估计程序以上交所58个周截面数据为样本进行实证研究,结果表明所给出的三因子HJM模型设定形式具有相对平稳的指数衰减结构,可以准确一致地表示取样期间内的国债远期利率期限结构.Under the HJM framework, forward interest rate term structure can be decomposed into two component functions： one can represent the observed initial forward rate curve; the other is the dynamic evolution of forward rate. Because they have the same parameters, it can simplify the practical implementation of a class of HJM models which allow the parameters of the model to be estimated from the initial forward rate function. In order to accurately fit the shape of forward rate curve, the paper introduces another spread factor on the basis of the original two-factor HJM model and derives a simple and accurate estimation procedure by utilizing the functionality of the extended HJM model. According to the procedure, the paper empirically investigates the ability of the model to fit the forward interest rate term structure by a sample of 58 weekly bond price data of Shanghai Stock Exchange. The results show that the three-factor HJM specification has stable exponential decay structure and is a consistent representation of the term structure of interest rate during the sampling period.

关 键 词：远期利率 分解技术 HJM模型 参数估计 遗传算法 

分 类 号：F830[经济管理—金融学]