基于logistic模型族的不良贷款极端零回收强度模型研究

不良贷款能否有回收是其定价、日常管理和回收策略的决定因素之一,而宏观经济和处置时效则是影响不良贷款能否有回收的双重利刃。本文依据我国最大的不良贷款数据库--LossMetrics^TM数据库,利用logistic模型族对清收时间跨度为2001-2008年的不良贷款零回收强度的动态变化影响因素进行了研究,并在研究中针对不同的样本分别建立了子模型和全模型,对多个模型的结果进行了对比。在全时间跨度模型中分析了GDP增速与零回收强度的直接关系;并把单笔贷款回收处置时间跨度分为小于12个月、12-22个月、23-60个月和超过60个月四组子样本,针对子样本分别建立模型,分析影响其各自零回收强度因素的区别。结果表明:GDP增速在大部分模型与零回收强度为显著负相关关系;在大部分子模型中不良贷款的有效抵质押因素显著,但在不同处置时间的子模型中显著情况有所不同。通过对零回收强度的研究可更好的结合宏观经济和处置时间来制订有效科学的回收策略。

Extreme Zero Recovery Rate Analysis Based on Family of Logistic Models

Whether the non-performing loan is zero recovery rate or not is one of the most important things for its pricing,management and disposal strategy.Macroeconomic factors and disposal effects will both affect the recovery rate.Based on LossMetrics ^TM,this paper use family of logistic models to analyze the hazard rate of zero recovery non-performing loans from 2001 to 2008.We analyze the relationship between macroeconomic factors and the zero recovery loans for the whole sample,and we divide the time of disposition into four subsamples,that is less than 12 months,12-22 months,23-60 months and more than 60 months,and contribute models for the four different subsamples.The results show that the relationship between GDP growth and zero recovery rate is negative,collateral factors are significant in most models for subsamples,but in the models the effects are different.Analyzing the relationship between macroeconomic factors and disposal effect will help us to develop effective recovery strategy