遗传算法选择性集成多分类器的企业财务困境预测

Financial Distress Prediction Based on Genetic Algorithm Selective Ensemble Multiple Classifiers

为了更加有效地进行企业财务困境预测,本文提出了基于遗传算法选择性集成的多分类器系统。与事先静态给定系统内部基本分类器组成不同,该方法以组合系统预测准确率为优化目标,无需度量各基本分类器之间的差异性,可以动态挖掘最优组合系统。实证研究中以中国上市公司为研究对象,以10折交叉验证准确率作为评价标准,结果表明该方法显著优于全集成以及单分类器最优模型。

In order to predict enterprises financial distress effectively,this paper puts forward a selective ensemble multiple classifiers system based on genetic algorithm.This method sets the predictive accuracy of the combined system as its optimization objective,needless to calculate the diversity between the base classifiers,which is different from traditional method that fixed its components in advance.Using Chinese listed companies data and adopting ten fold cross validation accuracy as an evaluation criteri...