基于logistic模型族的不良贷款极端零回收强度模型研究

发布时间：2018-06-04 23:04

本文选题：不良贷款 + 零回收强度　；参考：《中国管理科学》2011年02期

【摘要】：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——LossMetricsTM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。
[Abstract]:Whether bad loans can be reclaimed is one of the determinants of their pricing, daily management and recovery strategy, while macroeconomic and disposal limitation are the double blades that influence the recovery of non-performing loans. Based on the largest non-performing loan database in China, LossMetrics TM database, this paper studies the dynamic change factors of zero recovery strength of non-performing loans (NPLs) whose collection time span is 2001-2008 by using logistic model family. In the study, submodels and full models are established for different samples, and the results of multiple models are compared. The direct relationship between the GDP growth rate and the zero recovery intensity is analyzed in the full time span model, and the time span of a single loan recovery and disposal is divided into four sub-samples, which are less than 12 months, 12 to 22 months, 23 to 60 months, and more than 60 months, The model is established for the sub-samples, and the difference of the factors affecting their zero recovery strength is analyzed. The results show that the growth rate of GDP is negatively correlated with zero recovery intensity in most models, and the effective collateralization of non-performing loans is significant in most submodels, but it is different in submodels with different disposal times. Through the research of zero recovery intensity, we can better combine macroeconomic and disposal time to formulate effective and scientific recovery strategy.
【作者单位】：中央财经大学管理科学与工程学院;中国科学院数学与系统科学研究院;
【基金】：国家重点基础研究发展规划项目(2007CB814902) 国家基金委海外杰出青年基金(10628104);国家基金委创新研究群体(10721101) 国家水利部公益性项目(200801027) 中央财经大学“211工程”三期资助项目;中央财经大学青年科研创新团队基金资助项目
【分类号】：F832.4

【共引文献】