利率期限结构的长期记忆性及其预测作用研究  

Long-Term Memory Effect on the Term Structure of Interest Rate Forecasting

作　　者：纪哲翰 谢海滨[1] JI Zhehan;XIE Haibin(School of Banking and Finance,University of International Business and Economics,Beijing 100029)

机构地区：[1]对外经济贸易大学金融学院,北京100029

出　　处：《系统科学与数学》2023年第10期2598-2614,共17页Journal of Systems Science and Mathematical Sciences

基　　金：对外经济贸易大学中央高校基本科研业务费专项资金(19YB26)资助课题。

摘　　要：长期记忆性是时间序列的重要特征之一,对时间序列数据的预测有着重要影响.文章利用R/S分析、ARFIMA模型对中国利率期限结构的长期记忆性进行了实证分析,结果表明:中国利率期限结构存在显著的长期记忆性,刻画长期记忆性的ARFIMA模型可以显著地提升对利率期限结构的预测效果,并且随着预测步长的增加,预测效果提升得更为明显.因此,在实践中应重视利率期限结构的长期记忆性,充分发挥利率期限结构长期记忆性在利率期限结构预测中的作用.Long-term memory is one of the most important characteristics of time series and has great impact on time series forecasting.In this paper,based on R/S analysis and ARFIMA model,we perform an empirical analysis on the long-term memory of the term structure of interest rate of China.The results show that China's term structure of interest rate has significant long-term memory.Out-of-sample forecasting results show that ARFIMA model,which imports long-term memory infuence,significantly improves the forecasting of the term structure of interest rate,and the improvement becomes more obvious with the forecast step grows.Therefore,in practice,we should pay attention to the long-term memory of the term structure of interest and give full play to the role of long-term memory of interest rate term structure in the prediction of the term structure of interest rate.

关 键 词：利率期限结构 长期记忆性 R/S分析 ARFIMA模型 

分 类 号：F832[经济管理—金融学]