<p>In recent years, financial market instability has gradually increased. To better assist regulatory agencies and investors in assessing the credit risk of listed financial enterprises, this research proposed a credit risk measurement model for financial enterprises based on an optimized Harris Hawk optimization algorithm. Firstly, a credit risk assessment index for financial enterprises based on the optimized Harris Hawk optimization algorithm was established. Then, the parameters of the support vector machine algorithm were optimized. Finally, a credit risk measurement model for financial enterprises was constructed. The experiment outcomes indicated that the accuracy of the financial enterprise credit risk measurement model based on improved Harris Hawk optimization proposed by the research was 96.13%, the recall rate was 92.49%, and the F1 value was 96.82%. This indicated that the proposed model has good performance and can be utilized to credit risk evaluation of listed companies, which is helpful for regulatory agencies and investors in the financial market to make relevant decisions.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Credit risk measurement model of financial enterprises based on improved HHO

  • Hanqi Yang

摘要

In recent years, financial market instability has gradually increased. To better assist regulatory agencies and investors in assessing the credit risk of listed financial enterprises, this research proposed a credit risk measurement model for financial enterprises based on an optimized Harris Hawk optimization algorithm. Firstly, a credit risk assessment index for financial enterprises based on the optimized Harris Hawk optimization algorithm was established. Then, the parameters of the support vector machine algorithm were optimized. Finally, a credit risk measurement model for financial enterprises was constructed. The experiment outcomes indicated that the accuracy of the financial enterprise credit risk measurement model based on improved Harris Hawk optimization proposed by the research was 96.13%, the recall rate was 92.49%, and the F1 value was 96.82%. This indicated that the proposed model has good performance and can be utilized to credit risk evaluation of listed companies, which is helpful for regulatory agencies and investors in the financial market to make relevant decisions.