中国宏观杠杆率的水平测算与时变特征——基于MS-AR模型的实证研究  

Level Measurement of China′s Macro Leverage Ratio and Its Time-varying Features: An Empirical Study Based on the MS-AR Model

作　　者：李正辉 钟俊豪 董浩 LI Zhenghui;ZHONG Junhao;DONG Hao

机构地区：[1]广州大学金融研究院 [2]广州大学经济与统计学院

出　　处：《广州大学学报（社会科学版）》2019年第4期81-89,共9页Journal of Guangzhou University:Social Science Edition

基　　金：教育部人文社会科学研究规划基金项目(17YJAZH049)

摘　　要：不同视角采用不同方法测算宏观杠杆率,而最优的宏观杠杆率测算方法应与杠杆的经济目标具有高度相关性。文章利用2002—2017年的月度数据,运用Copula函数建模遴选出宏观杠杆率最优测算方法,发现货币供应量/GDP为样本期间内宏观杠杆率的最优测算方法。文章同时利用MS-AR模型,进一步分析了中国宏观杠杆率的时变特征,结果发现:一,中国宏观杠杆率具有显著的阶段性特征,且与具体经济事件和宏观经济政策有很强的相关性;二,中国宏观杠杆率可划分为高、中、低三个不同的区制,具有非线性的区制特征;三,中国宏观杠杆率在时变过程中具有非对称性。Different methods are adopted to measure the macro leverage ratio from different perspectives. The optimal macro leverage ratio measurement method should have a high correlation with the economic target of leverage. Based on the monthly data from 2002 to 2017, the paper uses Copula function modeling to select the optimal measurement method of macro leverage, and finds that the money supply/GDP is the best measurement method of macro leverage in the sample period. At the same time, using MS-AR model to further analyze the time-varying characteristics of China′s macro leverage ratio, this article finds that (1) China′s macro leverage has significant periodic features and is highly correlated with specific economic events and macroeconomic policies;(2) China′s macro leverage ratio can be divided into high, medium and low systems, showing nonlinear regional characteristics;(3) China′s macro leverage ratio is asymmetric in the process of time-varying.

关 键 词：中国宏观杠杆率 时变特征 COPULA MS-AR模型 

分 类 号：G203[文化科学—传播学]