商业银行小企业担保模式对信贷规模影响研究——基于LARS-Lasso算法  

Research on the Impact of Small Business Guarantee Model of Commercial Banks on Credit Scale: Based on LARS-Lasso

作　　者：方智 

机构地区：[1]江西财经大学统计学院,江西南昌330013

出　　处：《金融理论与实践》2018年第8期64-70,共7页Financial Theory and Practice

基　　金：2017年国家社会科学基金重大招标项目"新经济的统计测度研究"(17CTJ008)

摘　　要：基于小企业融资难的现实,利用A商业银行的小企业信贷数据,运用LARS-Lasso算法的回归模型分析小企业信贷担保模式对信贷规模的影响。研究结果表明:LARS-Lasso算法能够对高维数据进行高效降维和剔除多重共线性问题;保证担保不良率、抵押和质押担保放款额、保证和质押担保收回量对信贷规模成正影响;保证金担保放款额、信用和保证金担保收回量对信贷规模成负影响;抵押担保模式对信贷规模影响力在下降,而其他担保模式的影响力在逐渐上升。Based on the reality of small enterprise financing,using the small business credit data of A commercial bank,this paper uses the regression model of LARS-Lasso to analyze the impact of small business credit guarantee mode on credit scale.The results show that the LARS-Lasso can effectively reduce and eliminate the multiple collinearity of high dimensional data,and ensure that the guarantee rate,mortgage and pledge amount,guarantee and the amount of pledge return have a positive impact on the scale of credit.The influence of mortgage guarantee mode on credit scale is decreasing,while the influence of other guarantee modes is gradually increasing.

关 键 词：商业银行 信贷规模 融资难 担保模式 LARS-Lasso 

分 类 号：F830.33[经济管理—金融学]