考虑广义多维空间效应的S-VaR测算

S-VaR calculation considering generalized multidimensional space effect

VaR(value at risk)的测算精度一直是业界和学术关注热点问题.本文应用定义的经济测度距离和引力空间权重矩阵,建立广义多维空间计量模型捕捉金融系统的空间效应信息,构造S-VaR(saptial-value at risk),提高VaR的测算精度.以S&P亚洲50指数作为股票资产组合替代变量进行实证分析,结果表明:广义多维空间效应S-VaR能捕捉金融市场存在的多维空间相关性和风险的空间溢出效应,提高了VaR模型在风险预测中的精确性.

Calculation accuracy of VaR has been a hot issue in the field of the industry and the academic.This paper first defines economic metric distance and gravitational space weight matrix,establishes generalized multidimensional space econometric model to capture space effect information,and constructs spatial value at risk（S-VaR） model to improve VaR calculation accuracy.Last,we use SP Asia 50 index as the stock portfolio of alternative variables to test this model.The empirical results show that S-VaR can capture spatial correlation and risk spillover effect in financial market,and it raised the accuracy of VaR risk prediction.