A Comparison Study on Interest Rate Models of SHIBOR Based on MCMC Method

The main goal of this paper is to investigate the presence of jumps in Shanghai inter-bank offered rate (SHIBOR), which is Chinese money market benchmark interest rate, and compare interest rate models of SHIBOR based on MCMC method. Although SHIBOR has become an important interest rate, on which a lot of derivatives underlie, it is less studied. The Markov Chain Monte Carlo method is applied to analyze the interest rate models of SHIBOR, such as Vasicek model, Cox-Ingersoll-Ross model, CKLS model and CKLS jump diffusion model. The empirical results indicate that the CKLS model with generalized specification of volatility parameter is better than Cox-Ingersoll-Ross model and Vasicek model, but all these models are miss-specified. After introducing the jump factor, the model captures the jumps of 1-week SHIBOR rate well and passes the specification test. The estimates indicate the jump happens with a high probability everyday in the time period researched.