基于分形KMV模型的地方政府债券安全发行规模测度

Measurement on Safe Issuance Scale of Local Government Bond Based on Fractal KMV Model

文章构建分形KMV模型以测度31个省份地方政府债券的安全发行规模。首先验证31个省份财政收入的长记忆性,建立分形KMV模型;然后利用GM(1,1)与回归模型得到财政收入预测值,求解可担保财政收入;最后代入分形KMV模型,得出31个省份不同债券规模下的违约概率与安全发行规模。结果表明:若违约安全线p<0.4%,当发债规模为55%时,只有3个省份违约;当规模为65%时,几乎50%的省份违约;一旦规模超过85%,违约省份比例迅速提升至90%。

This paper constructs a fractal KMV model to measure the safe issuance scale of local government bonds in 31 provinces.Firstly,the paper verifies the long memory of the fiscal revenue in 31 provinces and constructs a fractal KMV model,then uses GM(1,1)and regression model to get the predicted value of fiscal revenue,and solve the guaranteed fiscal revenue,and finally substitutes it in the fractal KMV model to obtain the default probability and safe issuance scale of 31 provinces under different bond scale.The results show that in the case of default safety line p<0.4%,when the bond issuance scale is 55%,only 3 provinces default,that when the scale is 65%,almost 50%of the provinces default,and that once the scale exceeds 85%,the proportion of defaulting provinces rises rapidly to 90%.