Research On Green Credit Credit Risk Assessment And Pricing Strategy

In recent years,we have paid more and more attention to the construction of ecological civilization and green development,and promoting the high-quality development of green finance has elevated to national strategy.Relevant data shows that by the end of 2020,the accumulated balance of green credit in China was 904.5197 billion yuan,an increase of 67.15%compared with 2017,and exploring the way of green credit development is a significant part of the high-quality development of green finance.Green credit is a financial product that achieves reasonable allocation of environmental resources with the help of financial leverage.It is different from traditional credit in that its risk sources include potential environmental risks in addition to enterprise business risks,so the thesis will explore how to quantify green credit risks and achieve a reasonable allocation of environmental resources through differentiated pricing.Firstly,based on the traditional credit risk assessment theory,the thesis adds environmental risk refinement indicators and formulates a green credit risk assessment index system with 35 indicators including environmental characteristics.On this basis,the data of listed companies in China from 2010 to 2019 were used as data sources,the training set and test set are split,the secondary screening of indicators is performed using the recursive feature censoring method,and the risk assessment models based on LDA,Bayesian,XGboost and random forest are constructed in turn after using the ROSE algorithm to balance positive and negative samples.By comparing the performance of each model on the test set,it was found that the random forest model using the ROSE algorithm performs best in the evaluation indexes of AUC,Accuracy,and Precision,and the model AUC reaches 0.933,which can better identify high-risk enterprises and reasonably quantify green credit risk.Secondly,a theoretical study of common credit pricing methods was conducted,and the applicability of the RAROC model to the green credit pricing problem in China is demonstrated in light of the current situation.On this basis,the formulae underlying the RAROC model are derived,and a link between risk assessment and loan pricing is established by building a RAROC model that takes into account the cost of environmental risk.The model can achieve differentiated pricing of green credit by quantifying the environmental risk cost and promoting the rational allocation of environmental resources.Finally,a sample of real enterprises in green industries and high pollution industries are selected for the empirical study.The green credit risk assessment model is used to predict the default probability of borrowing enterprises,and the pricing model is used to calculate the sample interest rate and then compare it with the actual interest rate to test the overall model effect.Through the empirical study,it is found that the random forest model using the ROSE algorithm can predict the probability of green credit default by combining enterprise business risk and environmental risk,and then effectively identify high-risk enterprises.The RAROC model with environmental risk cost can dynamically adjust the lending rate according to the environmental impact of enterprises,so that the capital flows to low-risk green enterprises,and promote the green and sustainable development of banks and enterprises.The conclusions of the thesis can provide a reference for green credit risk assessment and pricing strategy,and provide a theoretical basis for the realization of refined management in the context of green credit scaling in China.