Study On Price Duration Of Chinese Bond Market

Bond market, as an important carrier of national monetary policy, is always an essential component of financial market. Unlike the OTC trading in most of the other countries, Chinese exchange bond market adapts trading via electronic facilities, and in this way the exchange trading can provide less transaction cost as well as the high-frequency data which contain finer market trading information. Therefore, it is of significance to study the structure of Chinese exchange bond market. Moreover, as a new research field of financial market, the study on durations based on high-frequency data also has economic significance because the durations embody some trading information like intensity and liquidity. The main content of this paper is to model the price durations using the high-frequency bond trading data from Shanghai exchange, with the econometric method of financial high-frequency series and autoregressive conditional duration model, in order to characterize and explore the dynamic behavior of price durations in Chinese exchange bond market, which is considerable information for market participants.The research data in the paper include high-frequency trading data of 7 bonds in 2003.And it can be concluded from the result of modeling and statistics analysis that: the bond price duration series in Shanghai Exchange exhibit a inverted U shape diurnal pattern and serial correlation; by setting the model flexibly, ACD model can explain and characterize the dynamic mode of bond price duration series as some bonds price duration exhibit a strong cluster effect while some other bonds do not show significant cluster effect, and the Chinese bond exchange market was becoming active in the frequency of price changes according to the estimate result of a monotonously decreasing hazard function. These results are different from those of stock market where the cluster effect is ubiquitous and the hazard function is monotonously increasing.