Modeling the Dynamics of Chinese Spot Interest Rates

Understanding the dynamics of spot interest rates is important for derivatives pricing, risk management, interest rate liberalization, and macroeconomic control. Based on a daily data of Chinese 7-day repo rates from July 22, 1996 to August 26, 2004, we estimate and test a variety of popular spot rate models, including single factor diffusion, GARCH, Markov regime switching and jump diffusion models, to examine how well they can capture the dynamics of the Chinese spot rates and whether the dynamics of the Chinese spot rates has similar features to that of the U.S. spot rates. A robust M-estimation method and a robust Hellinger metric-based specification test are used to alleviate the impact of frequent extreme observations in the Chinese interest rate data, which are mainly due to IPO. We document that GARCH, regime switching and jump diffusion models can capture some important features of the dynamics of the Chinese spot rates, but all models under study are overwhelmingly rejected. We further explore possible sources of model misspecification using some diagnostic tests. This provides useful information for future improvement on modeling the dynamics of the Chinese spot rates.