Enhancing the Fruit Fly Algorithm for Z-Score Financial Early Warning Models

Abstract

The purpose of this study was to enhance the accuracy of financial data prediction and decrease the root mean square error (RMSE) by improving the FOA algorithm and introducing the YJFOA Z-score model. The financial data of 29 ST companies (including *ST) listed on the Shanghai Stock Exchange and Shenzhen Stock Exchange and 29 non-ST companies in the Yangtze River Delta from 2015 to 2019 were analyzed using the optimized Z-score model based on the new YJFOA-ZSCORE model. By enhancing the Fruit Fly algorithm’s performance, the study discovered the limitation of the traditional Z-score model in evaluating the financial situations of Chinese A-share listed companies. As a result, a suitable threshold was identified to create a new financial early warning system. The study concluded that the new YJFOA Z-score model outperformed the FOA-ZSCORE and PSO-ZSCORE models in predicting all types of companies.