摘要
本文综合宏观经济变量,利用混频动态因子模型测度月度GDP,之后结合货币供应量、房价以及股价等金融变量,基于贝叶斯时变VAR模型构建科学的金融状况指数(FCI),并采用谱分析研究其预警能力.研究结果表明,股票市场、房地产市场、货币政策与实体经济对我国金融市场状况的影响具备时变性.基于贝叶斯时变VAR模型的脉冲响应分析可以发现,上述变量的影响程度依次递减.在此基础上构建的FCI较基于常系数VAR模型构建的FCI具备更强的预警能力,尤其进行三个月以内的短期预测时,优势更加明显,进而可以为有效监测金融市场状况,预警通货膨胀与金融风险提供科学决策依据.
This paper combines with macroeconomic variables,the monthly GDP is measured using a mixing dynamic factor model.After that,combined with financial variables such as money supply,house prices,and stock prices,FCI is constructed based on the Bayesian time-varying VAR model,and its early warning capability is studied using the spectrum analysis.This paper finds the impact of stock market,real estate market,monetary policy and substantial economy on the financial market condition in China bears time-varying characteristics.It is found in the pulse response analysis of Bayesian time-varying VAR model that the impact of the above variables gradually decreases.The FCI built in this paper is of stronger early warning capability compared that built on the constant coefficient VAR model and this advantage is more significant when it comes to the short-term forecast within three months,then provides a scientific decision-making basis for the effective monitoring of financial market conditions,early warning of inflation and financial risks.
作者
肖强
汪卢俊
XIAO Qiang;WANG Lu-jun(School of Statistics,Lanzhou University of Finance and Economics,Lanzhou 730020,China;School of Finance and Taxation,Nanjing University of Finance and Economics,Nanjing 210023,China)
出处
《数理统计与管理》
CSSCI
北大核心
2023年第2期191-204,共14页
Journal of Applied Statistics and Management
基金
国家自然科学基金(71763016,72063022)
国家社科基金(19FJYB011)。
关键词
金融状况指数
混频动态因子模型
贝叶斯时变VAR模型
谱分析
financial condition index
mixed dynamic factor model
Bayesian time-varying VAR model
spectral analysis
