基于信贷风险量化分析的辅助决策模型

Auxiliary Decision-Making Model Based on Quantitative Analysis of Credit Risk

本文对中小微企业的信贷决策问题进行讨论,从企业相关数据中提炼出多项指标,利用其建立量化分析模型,并结合如信誉评级等因素制定了信贷策略。首先,使用主成分分析方法对指标进行降维处理,并采用逻辑回归算法对企业的违约概率进行预测,用企业违约概率量化了企业的信贷风险;其次,引入信誉评级作为新的指标,并定义惩罚因子来更全面地约束模型;最后,计算得出各企业贷款对各企业贷款年利率、贷款配额权重的方案,并与市场实际情况进行对比.对比结果表明,该方案符合市场规律,具有较强的可靠性.

This paper discusses the credit decision-making of small and medium-sized and micro enterprises,extracts a number of indicators from the relevant data of enterprises,establishes quantitative analysis model by using them,and establishes Credit Strategies Based on factors such as credit rating.Firstly,the principal component analysis method is used to reduce the dimension of the index,and the probability of default is predicted by the logic regression algorithm,and the credit risk of the enterprise is quantified by the probability of default;Secondly,credit rating is introduced as a new index,and penalty factors are defined to constrain the model more comprehensively;Finally,the paper calculates the scheme of annual interest rate and quota weight of each enterprise loan,and compares with the actual situation of the market.The comparison results show that the scheme is in line with the market law and has strong reliability.