摘要
本文利用上海证券交易所1996年7月22日至2004年8月26日间的7天国债回购利率对各种短期利率模型进行了实证分析和检验。结果表明,引入GARCH、机制转换以及跳跃因子可大大地提高短期利率动态模型的拟合效果。我们发现在1996-1998年间,国债回购利率水平、波动性以及突然跳跃的概率都要高于1999年以后,但是利率波动性对利率水平变化的敏感性则在1999年以后变得更强了。我们使用了Hong and Li(2004)新近提出的非参数检验方法比较了各个模型的相对表现,发现没有一个模型可以准确描述中国短期利率波动。
We examine a wide variety of popular spot interest rate models in China using daily data of 7-day repo rates from July 22, 1996 to August 26, 2004. The estimation suggests that introducing GARCH, regime-switching and jump effect substantially improves the fitting degree of the spot rate models. We find that the level and volatility of the spot rates, and the jump probability were significantly higher before 1999. However, the sensitivity of interest rate volatility to the change of interest rate level became stronger after 1999. To further compare the relative performance of different models, we use the nonparametric specification test recently proposed by Hong and Li (2004). The results show that we still have not obtained an exact dynamic model.
出处
《经济学(季刊)》
2006年第2期511-532,共22页
China Economic Quarterly
基金
中国国家自然科学基金会海外杰出青年基金
教育部“长江学者讲座教授”计划的资助
厦门大学王亚南经济研究院“2005年科研计划”的资助