摘要
以国债回购利率为研究对象,分别建立ARIMA及GARCH模型,并比较这两种模型的预测能力。研究结果表明:使用传统ARIMA模型,模型ARIMA(0,1,1)配适较好;使用GARCH模型,模型GARCH(2,3)配适效果较好。此外,虽然GARCH模型的预测置信区间的波动性比ARIMA模型要小,但ARIMA模型的预测置信区间更小一些,因此其预测能力比GARCH模型更强。
The repo rate of the national bond is analyzed and ARIMA and GARCH models related to the rate are established in this paper. Through comparing the predicting capability of the two models, the one which is more suitable for predicting the repo rate is chosen.
出处
《统计与信息论坛》
CSSCI
2008年第5期28-31,共4页
Journal of Statistics and Information