Research On Credit Scoring And Pricing Of Rural Household Microfinance Guarantee

In recent years, the problem of “Agriculture, Rural and Farmers” has been the most important task of China’s economic development. Practice has proved that, microfinance can provide effective financial support for agricultural economic. Although farmers are playing an increasingly important role in the development of today’s society, but in the process of its development, lacking of credit and financing difficulty become the main obstacles. In this context, rural household microfinance guarantee came into being, with the introduction of the policy support, which has got extensive attention and application in many cities and provinces of China. In China, rural household microfinance guarantee is still a new thing, the play of its function and sustainable development relates to a series of problems such as the operating system structure, management experience, price decision, risk sharing and avoiding, the core of the problems are the assessment of credit guarantee business risk and pricing decision. Farmers own the characteristic of high risk which is especially obvious, if can not effectively identify the risk of guarantee business, guarantee institutions are likely to become a carrier of credit risk, which leading to a difficult dilemma. Therefore, it is very important to strengthen the professional credit risk control and improve the bargaining power of the guarantee institutions.This paper analyzes the influence factors of farmers credit from qualitative and quantitative angel, and combines with the characteristics of farmers themselves to analyze the evaluation methods of farmers credit systematically, then chooses the BP neural network model as a basic credit evaluation model and expounds the advantages and limitations of the BP neural network through the analysis of theory and algorithm. This paper improves the model from two aspects: In order to avoid network training caught in shallow local minima increases momentum; in order to guarantee the convergence of the algorithm to accelerate the convergence speed, adopt the strategy of learning rate self-adjustment. On the basis of summarizing and comparing the pricing model of credit guarantee, and combines with the characteristics of farmer credit guarantee business to adjust the method of RAROC, establishing a microfinance guarantee pricing model which considers to make different guarantee fees according to different credit rating and collateral value, and strives to establish a simple, scientific and practical microfinance guarantee business pricing method.