Empirical Study of Nonparametric Model of Interest Rate Term Structure Based on Different Kernel Functions

The term structure models of interest rate have been extensively applied to asset pricing, design of financial instruments, hedging, arbitraging and investment decision. Therefore, the estimation of parameters in the term structure model has been a key problem. In this paper, the parameters of term structure model is estimated by using two different kernel functions: Gauss kernel function and Epanechnikov kernel function with the data of the repurchasing rate in Shanghai stock market. The empirical results show that the density function of short interest rate is non-normal distribution, and both the drift and diffusion function are nonlinear, which is accord with the facts of financial market in China. Meanwhile we find that the estimation effects of Gauss kernel function are almost identical with one of Epanechnikov kernel function.