非参数利率期限结构模型的实证检验

利用随机过程的原理和非参数的核函数估计方法建立了非参数的利率期限结构模型，采用上海证券交易所的国债回购利率数据对建立的模型进行了实证检验，并与Vasicek、CIR模型进行了比较．结果显示，短期利率扩散过程的漂移函数和扩散函数都是非线性的，短期利率的概率分布不服从参数模型所假设的分布．实证结果证实，与参数模型相比，非参数模型能够更准确地描述我国的国债利率期限结构状况．

Empirical Testing on Nonparametric Model of Interest Rate Term Structure

One nonparametric model was established by use of the stochastic theory and kernel estimation method. By use of the data of the repurchasing rate in Shanghai stock market, the nonparametric model was empirically tested compared to Vasicek and CIR. The testing result discloses that the drift and diffusion function are nonlinear, and the distribution of interest rate does not follow the specification in the parametric models. So the nonparametric model fits the interest rate term structure better than the parametric models.