摘要
使用累积分布函数-信用加权方法构建反映系统性金融风险累积状况的压力指数,并引入动态模型平均方法对该指数走势进行预测分析并评估其预测效果.结果表明,动态模型平均方法在模型选择、变量选择、系数选择等方面具备良好的灵活性,具有良好的预测效果.在具体应用中,GDP增长率与CPI增长率是对金融压力指数预测最有帮助的一类指标,其在后期的包含概率显著持续高于0.5,这类指标在风险防范中更值得关注.
This paper used systemic financial stress index which reflect the accumulation status of systemic financial risk based on the cumulative distribution function- weighted credit method,and introduced the dynamic model of average method to estimate the index trend and prove its predictive effect. The results showed that he dynamic model of average method had good flexibility and effect in the respects of model selection,variable selection,coefficient selection and so on. In the specific application,the GDP growth rate and CPI growth rate were the most helpful index for the forecasting of systemic financial stress index（ SFSI）,the coverage probabilities in the following periods were always above 0. 5,which should be pay more attention in the risk prevention.
出处
《哈尔滨商业大学学报（自然科学版）》
CAS
2016年第1期107-111,共5页
Journal of Harbin University of Commerce：Natural Sciences Edition
基金
教育部人文社会科学重点研究基地重大项目(14JJD790028)
教育部人文社会科学研究青年基金项目(13YJC790122)
辽宁省社会科学规划基金一般项目(L13CJL027)
辽宁省高等学校杰出青年学者成长计划(WJQ2015001)
关键词
系统性金融压力指数
动态模型平均
预警
systemic financial stress index
dynamic model of average
early warning
