人民币汇率市场化,结构相依与结构突变  

Marketization,Structural Dependence and Structural Break of RMB Exchange Rates

作　　者：吴恒煜[1,2] 胡根华[3,2] 吕江林[4] 

机构地区：[1]暨南大学管理学院,广东广州510632 [2]西南财经大学经济信息工程学院,四川成都611130 [3]安徽工业大学商学院,安徽马鞍山243032 [4]江西财经大学金融学院,江西南昌330013

出　　处：《数理统计与管理》2016年第1期106-121,共16页Journal of Applied Statistics and Management

基　　金：国家自然科学基金重大研究计划(91218301);国家自然科学基金面上项目(71171168);教育部社科研究基金规划项目(13YJA790076,13YJA790104);2014年教育部人文社会科学研究西部和边疆地区项目(14XJA790002);国家留学基金(201406980031);中央高校基本科研业务费专项资金项目(JBK1407165,JBK1507K09,JBK150504);重庆市社会科学规划博士项目(2014BS026);奥胡斯大学时间序列计量经济分析研究中心(CREATES)丹麦国家研究基金(DNRF78)的资助

摘　　要：运用能够识别资本市场结构突变与区制变化的Markov区制转换模型,基于非线性相依结构研究中的藤Copula分析框架,文章以考察人民币汇率市场化进程中的结构相依与突变特征为切入点,重点研究两次汇改以及金融危机时期人民币汇率在四个阶段的结构转换及非对称动态相依特征。文章采用GJR-GARCH模型探讨人民币汇率市场的"杠杆效应"。在此基础上,文章对两次汇改以及美国次贷危机时期人民币汇率市场的结构突变和区制转换的进行识别。研究发现,Markov区制转换模型能够准确地捕捉到人民币汇率第一次汇改的临界点,但在捕捉第二次汇改临界点方面却存在一定的滞后反应。并且,该模型对美联储采取第一轮量化宽松的货币政策的捕获,也表现出较好的能力。进一步地,文章运用藤Copula分析框架探讨了不同人民币汇率市场之间的非线性相依结构。研究表明,整体而言,采用t-Copula的藤结构在捕捉人民币汇市之间的相依结构方面表现出良好的刻画效果。Employing Markov regime switching model to recognize the structure break and regime switching of capital markets, based on vine Copulas analytical framework for capturing asymmetric dynamic dependence, the present research is directly to analyze the asymmetric dynamic dependence and structure switching of RMB exchange rates during the marketization process, using Markov regime switching and vine Copulas models in four different stages before and after two RMB exchange rate reforms and 2008 financial crisis. It firstly models GJR-GARCH structure to capture the leverage effects among the RMB exchange rate markets. On that basis, it this paper tries to recognize the structure break and regime switching of RMB exchange rate markets. It is found that Markov regime switching model can perfectly capture the threshold of the first RMB exchange rate reforms, but not the second one. Besides, it displays good capture ability to the first round US Monetary policy of Quantitative Easing. Furthermore, the present research is to capture the asymmetric dynamic dependence structure among different RMB exchange rate markets. The empirical study shows that the dependence structures of RMB exchange rates can generally be described by the Vine Copulas with Student＇s t distribution.

关 键 词：汇率 结构相依 结构突变 MARKOV区制转换 藤Copula 

分 类 号：F830.9[经济管理—金融学] O212[理学—概率论与数理统计]