基于多分类有序Logit模型的农户信用等级影响因素  

Influencing factors of the credit rating based on the Polytomous Logit model

作　　者：王霞[1] 吕德宏[1] 

机构地区：[1]西北农林科技大学经管学院,陕西杨凌712100

出　　处：《中国农业大学学报》2013年第3期209-214,共6页Journal of China Agricultural University

摘　　要：针对农村金融机构内部农户信用评级问题,依据西部农村小额信贷农户的调查数据,采用多分类有序Logit模型对农户信用等级影响因素进行实证研究。结果表明:耕地面积、家用电器价值、家庭年总收入、贷款利率、地区经济水平、机构服务满意度对农户信用等级有显著影响;年龄、劳动力人数、教育水平、专业技能、职业、房屋价值、贷款金额、贷款期限、贷款用途、信誉状况对农户信用等级的影响不显著;显著因素的内部次序对农户信用等级的影响表现出明显的差异性。建议小额信贷机构对农户信用等级评定时应侧重以上显著影响因素,并考虑其内部次序所带来的差异,从而建立更加完善的针对不同等级农户的信用评级制度。Based on the survey of western rural microfinance farmers,the influencing factors on farmer＇s credit was analyzed using polytomous logit model in this paper.The result showed：arable land,household appliances value,the total annual household income,loan interest rate,the level of regional economic and institutional service satisfaction significantly affected farmers＇ credit rating.But,farmer＇s age,labor size,education level,professional skills,occupation,housing value,loan amount,loan period,loan purpose as well as credit history were not significantly associated with credit rating.In addition,the internal order of a significant factor revealed the differences of the impact on credit rating.The factors that significantly influenced credit should be taken as main factors into consideration and the internal order difference as additional ones for a reasonable credit rating system construction.

关 键 词：农户 信用等级 有序LOGIT模型 影响因素 显著差异 

分 类 号：F812.8[经济管理—财政学]