Study On The Evaluation Of Tea Farmers' Credit Risk

China is the largest tea planting country and the fastest growing country in the area of tea plantation in the world.Tea industry has become a pillar industry in some areas.However,as the bottleneck of the development of tea industry,the financing situation of tea farmers has always been severe.Credit rating plays a positive role in easing the constraints of farmers' financing,and helps customers obtain credit from the formal financial sector.At the same time,due to the poor availability of farmers' credit data,as well as their poor financial and non-financial information,farmers' credit risk is difficult to accurately describe,and there are few research on the dividing the whole group of farmers into specific industry.In view of this,this paper takes 1814 tea farmers in Anxi and Wuyishan of Fujian Province as examples,and constructs a credit risk evaluation index system that conforms to the characteristics of tea farmers by using the lasso logistic method.On this basis,through mining the key characteristics of the indicators that have an important impact on the credit risk of tea farmers,this paper puts forward countermeasures and suggestions to improve the credit status of tea farmers,in order to provide reference for the evaluation and management of credit risk of tea farmers.There are five chapters in this paper: the first chapter is the introduction,which mainly introduces the background and significance of the research on tea farmers' credit risk assessment,combs the relevant literature and summarizes the research content,ideas,and the potential innovation of this study.The second chapter is the theoretical basis of tea farmers' credit risk assessment.This chapter defines the definition of farmers,tea farmers,credit,credit risk and credit rating,and expounds the theoretical basis of credit risk assessment.The third chapter is the construction method of tea farmers' credit risk assessment.The credit risk assessment model of tea farmers includes the preprocessing of the original data of tea farmers' credit assessment,the selection of tea farmers' credit assessment indicators,the weighting of indicators,the solution of credit score,the classification of credit rating and the mining of key characteristics of indicators that have an important impact on tea farmers' credit risk.The fourth chapter is an empirical analysis of credit risk assessment based on 1814 samples of effective tea farmers in Anxi and Wuyishan.First of all,the lasso logistic method is used to deal with the sparse data of high dimension of tea farmers,and the redundant indicators of credit risk evaluation are eliminated,and the credit rating index system of tea farmers is constructed.Secondly,by using Fisher discriminant method,the weight of tea farmers' credit evaluation index is calculated,and then the credit scores of tea farmers are obtained.Thirdly,the fuzzy c-means method is used to classify the credit rating of tea farmers under different credit risks.Finally,based on the non-parametric test,mining which characteristic attribute of the same index is the key feature that affects the default of tea farmer loan.The fifth chapter is conclusion and suggestion.The empirical results are as follows: firstly,using the credit data of 1814 tea farmers in Anxi and Wuyishan,taking the default of tea farmers as the dependent variable,the lasso logistic model is constructed,and 8 factors that can distinguish the default of tea farmers are screened out,such as gender of tea farmers,supporting population,county classification,main income source of families,total debt of families,contract status,total expenditure of families during the expected loan period and risk classification Important indicators of status.Secondly,the results of credit rating of tea farmers show a normal distribution,46.58% of tea farmers are located in BBB,and the credit risk of tea industry is low as a whole.Finally,through the feature mining of eight important indicators that affect the default state of tea farmers,it is concluded that the default risk of tea farmers who are male and support population is 1 person,non-key counties,Wuyishan tea farmers,risk is classified as secondary,family debt is 200000-400000,and total family expenditure is more than 400000 is higher.The innovation of this paper is as follows: first,using the credit data of 1814 tea farmers in Anxi and Wuyishan,through lasso logistic quantitative screening,the paper constructs a credit risk evaluation index system of tea farmers,which is composed of eight indexes,i.e.tea farmers' gender,each tea farmers' supporting population,classification of tea farmers' counties,main income sources of tea farmers,total household liabilities,etc.Second,using non-parametric test,mining which characteristic attributes under the same index are the key features that affects the credit risk of tea farmers.This study method has important reference for financial institutions to identify tea farmers' loan customers and credit risk enhancement of tea farmers.