基于门限自回归模型的中国财政风险预警系统  

Early Warning System of Fiscal Risk in China Based on Threshold Autoregressive Model

作　　者：孟庆斌[1] 杨俊华[1] MENG Qing-bin YANG Jun-hua(School of Business, Renmin University of China, Beijing 100872)

机构地区：[1]中国人民大学商学院

出　　处：《中国人民大学学报》2016年第6期86-94,共9页Journal of Renmin University of China

基　　金：国家社会科学重大招标项目"中国财政金融安全:预警机制与风险控制体系"(05&ZD008);中国人民大学校内项目"卖空机制;股价信息效率与知情交易"(2015030083)

摘　　要：财政在整个宏观经济中占据着重要的地位,财政风险的出现往往对国民经济产生强烈冲击,甚至影响到社会的稳定。赤字风险是衡量财政风险程度的一个较好的尺度。在大部分时间里,我国赤字风险都较为平稳地处于中、低度风险状态。但在1999年到2003年的后东南亚金融危机时期,以及2008年之后的国际金融危机及经济转型时期,我国赤字风险程度都上升到了较高的水平。运用门限自回归模型可以较好地构建赤字风险预警系统,从而能够前瞻性地对赤字风险做出预警,采取相应对策化解或尽可能地降低财政风险。Based on the fiscal red figure,we investigate the fiscal risk between January 1994 and Sep-tember 2019 in China,and builds up an early warning system for fiscal crisis.Firstly,we define the in-dex of deficit risk,and give a qualitative analysis of the situation of deficit risk of China by means of HP filtering.Secondly,with Threshold Autoregressive Model,we get two threshold values and divide the fiscal risk situation of our country in the sample studied by this paper into high,low and middle risk levels.Finally,we divide the data sample into training and forecast parts,and builds up the deficit identification model and Early Warning System with each of them respectively.The result shows that, the deficit identification model performs well in measurement of the deficit risk,and the Early Warning System of deficit risk also gets a good score when carrying through the out-sample forecast.

关 键 词：财政风险预警 赤字风险指数 门限自回归模型 

分 类 号：F812[经济管理—财政学]