摘要
基于时变参数视角,拓展了原有的金融稳定状态指数构建方式,建立时变参数因子加强型向量自回归模型TVP-FAVAR,选取2004年第3季度至2014年第3季度的24类宏观经济数据作为样本进行实证研究。结果表明:改进的金融稳定状态指数能够有效预测未来5-7个季度内我国的宏观经济走势.未来l-3个季度内的通货膨胀趋势,以及未来6-8个季度的货币错配风险水平。该金融稳定状态指数包含未来较长一段时期内的金融变量信息,时变参数的优势使得该指数能够及时反映我国金融制度和结构的变化,有助于对我国经济政策实施的效果进行科学评估,从而便于政府对经济政策进行及时调整。
This paper extends the original FSCI construction methods, varying parameters factor when using TVP-FAVAR model (Time Varying Parameters Factor Augmented Vector Auto Regression), which enable us to take 24 kinds of macroeconomic indicators from the 3 quarter of 2004 to the 3 quarter of 2014 into the model. The empirical research shows: this improved FSC1 can effectively predict China's macroeconomic trend over next 5-7 quarters, the inflation trend over next 1-3 quarters and the risk of currency mismatch over next 6-8 quarters, which shows that the FSCI contains information about financial variables in the distant future. Advantage of time varying parameter also allows the index to reflect structural changes of the financial system, which will help to scientifically evaluate the effect of the imple- mentation of economic policies, and provide a strong basis for the government to make timely adjustments to monetary policy.
出处
《系统工程》
CSSCI
CSCD
北大核心
2016年第10期19-26,共8页
Systems Engineering
基金
国家社会科学基金资助项目(14BJY201)