农户小额贷款违约影响因素研究

发布时间：2018-01-15 11:09

本文关键词：农户小额贷款违约影响因素研究　出处：《西北农林科技大学学报(社会科学版)》2017年03期 　论文类型：期刊论文

【摘要】：基于江苏北部某地区2007—2015年18万条真实农户大样本小额贷款记录,结合当地及全国CPI、GDP、第一产业产值、地区农业生产资料价格指数等宏观经济指标,采用基于最大似然估计逐步进入法的Logistic模型,筛选对农户违约概率影响较为显著的指标,之后对各指标经济含义进行了解释并对模型进行稳健性分析。研究发现:信用水平指标与农户的真实违约情况关联不显著,意味着当地信贷机构对贷款农户的贷前内部信用评级不能有效地预测农户的信用风险;利率、性别、婚姻状况、职业、教育等微观指标对信用风险有较大影响;全国范围的GDP和CPI、江苏农业生产资料价格指数、滞后一期的当地第一产业GDP等宏观指标也对信用风险具有预测作用;Logistic模型在不平衡数据集上,依然能保持较好的分类精度。
[Abstract]:Based on the large sample microloan records of 180,000 real farmers from 2007 to 2015 in a certain area of northern Jiangsu, combined with local and national CPI GDPs, the output value of the primary industry was obtained. The Logistic model based on the maximum likelihood estimation (MLE) is used to screen the indexes which have a significant impact on the farmers' default probability, such as the price index of regional agricultural means of production, and so on. Then it explains the economic meaning of each index and analyzes the robustness of the model. It is found that the credit level index is not significantly related to the true default of farmers. This means that the internal credit rating of local credit institutions can not effectively predict the credit risk of farmers; Interest rate, gender, marital status, occupation, education and other micro-indicators have a greater impact on credit risk; The national GDP and CPI, the price index of agricultural means of production in Jiangsu Province, the local primary industry GDP and so on, also have the forecast function to the credit risk. Logistic model can still maintain good classification accuracy on unbalanced data sets.
【作者单位】：南京大学商学院;泰国国家发展管理学院;中国人民银行连云港中心支行;
【基金】：国家自然科学基金重大研究计划项目(90718008);国家自然科学基金青年项目(71403124) 国家社会科学基金重大项目(14ZDA043)
【分类号】：F832.4
【正文快照】： 3.中国人民银行连云港中心支行,江苏连云港222000)引言随着我国经济的不断发展以及一系列支农、惠农政策的扶持,农户的收入及生活水平不断提高,扩大生产的积极性与消费需求日益增长。然而,尽管当前农户对资金的需求在不断提高,农户信贷市场上却存在着严重的资金供求不平衡问题

【相似文献】