利率期限结构与宏观经济的区制依赖关联机制——基于宏观-金融MSV-DRA模型建模及实证  

The Regime-dependence Relevance between the Term Structure of Interest Rates and Macroeconomy——Based on MSV-DRA and Empirical Tests

作　　者：杨婉茜 成力为[2] YANG Wan-qian;CHENG Li-wei(School of Business,Nantong University,Nantong 226019,China;School of Economics, Dalian University of Technology, Dalian 116024, China)

机构地区：[1]南通大学,江苏南通226019 [2]大连理工大学,辽宁大连116024

出　　处：《系统工程》2018年第1期1-12,共12页Systems Engineering

摘　　要：将马尔科夫区制转移变量引入经典的DRA模型,构建并采用Kim滤波求解MSV-DRA模型,从宏观经济状态转变的视角研究中国银行间国债利率期限结构与宏观经济非线性动态关联的区制依赖特征得到:首先,模型拟合和区制依赖检验结果表明两区制MSV-DRA模型比单一区制DRA模型更适合描述收益率曲线与宏观经济的非线性关联性;第二,脉冲响应分析显示宏观-金融变量之间的冲击在不同区制下有较大区别,方差分解结果表明宏观经济因素在高波动区制下对收益率预测方差的解释比重高于在低波动区制下的解释比重,反映宏观变量和区制状态都是收益率预测的必要考虑因素。第三,通过Logit模型和泰勒规则对模型区制进行识别,发现其可以被分别理解为货币政策的宽松和紧缩状态,因此利率期限结构与宏观经济的动态关联表现出明显的货币政策状态依赖特征。We introduce a Markov regime-switching variable into DRA, then construct and solve a macro-financial MSV-DRA with Kim filter to study the state-dependence of the relevance between the term structure and macroeconomy from the perspective of state-shifting. And we get some findings as follows： first, MSV-DRA is more appropriate for China than DRA from two aspects of model fitting and state-dependence test. Second, the impacts between macroeconomic factors and financial variables is obvious different in two regimes, and the variance decomposition indicate that macroeconomic factors can explain more of yield prediction variance in high volatility than in low volatility, reflecting macroeconomic variables and the regime are all necessary for yield prediction. Third, we identify the two regimes as the loose- and tight-state of monetary policy.

关 键 词：利率期限结构 马尔科夫区制转移 宏观经济 宏观-金融模型 

分 类 号：F830[经济管理—金融学]