Analysis of Structural Breakpoints of China’s Bond Market and Important Economic Event Based on ICSS: MV Algorithm

Abstract

In this paper, we use the algorithm of Iterated Cumulative Sums of Squares for changes of Mean and Variance (ICSS: MV) to estimate structural breakpoints of yield series of China Securities Index (CSI) Universal Bond Index during the period from Jan., 2003 to May, 2010. We find there are 15 structural breakpoints. Having classified the type of each breakpoint, we find that most of them are variance changes. In all of the events that have effect on the volatility of China’s bond market, the events of adjusting reserve ratio and interest rate and the release of macroeconomic data account for a substantial portion. We also find there are not Treasury bond auction announcements in these significant events, the reason may be due to the different issuing ways of Treasury bond between China and the US. In addition, comparing the empirical results in this paper with the researches on structural breakpoints of China’s stock market, we find that the degree of influence caused by significant events on China’s bond market is less than that on stock market.