基于GA-SVR模型的中国上市公司融资风险预测

提升融资风险预测精度对促进企业资金融通与缓解企业融资约束问题具有重大意义。将中国上市公司作为研究对象，利用粗糙集理论（RS）筛选出企业融资风险预测指标，利用遗传算法（GA）对支持向量回归（SVR）模型的参数进行寻优，并采用GA-SVR模型预测上市公司的融资风险。研究表明:粗糙集理论筛选后得到的17个融资风险预测指标具有较强区分企业是否出现异常状况的能力；资产规模对上市公司的风险预测具有至关重要的作用；GA参数寻优后的SVR模型具有良好的预测精度与稳健性。

Financing Risk Prediction of China's Listed Company based on GA-SVR Model

Improving the prediction accuracy of corporate financing risk is of great significance for promoting corporate financing and alleviating corporate financing constraints.In this paper,the listed companies in China''s listed company are taken as the research object,the financing risk forecasting index is screened by using rough set theory(RS),the parameters of support vector regression(SVR)model are optimized by genetic algorithm(GA),and GA-SVR model predicts financing risks of listed companies.The research shows that the 17 financing risk forecasting indexes obtained by rough set theory screening have a strong ability of distinguishing whether an enterprise is abnormal or not; the scale of assets plays a crucial role in the risk prediction of listed companies; SVR model has good prediction accuracy and robustness.