基于遗传算法和熵的缩减记忆式LS-SVM财务困境预测模型研究

为了提高财务困境预测的正确率,减少模型的训练样本数和训练时间,在传统支持向量机(SVM)预测模型的基础上,将遗传算法、信息熵和缩减记忆算法应用于最小二乘支持向量机(LS-SVM),提出了一种基于遗传算法和信息熵的缩减记忆式最小二乘支持向量机预测模型。并独立推导出了适合财务困境预测这一离散序列的熵以及支持向量机核函数的表达式,同时,给出了这一改进模型的实现步骤。实验结果表明,该模型无论是预测正确率,还是训练样本的数量和训练时间,都显著优于最小二乘支持向量机以及传统支持向量机模型。

Study on Financial Distress of Prediction Model of Least Square Support Vector Machine of Memory-Reduced Algorithm Based on Genetic Algorithm and Entropy

In order to improve the accuracy of financial distress prediction and reduce the sample number and training time,this paper applies genetic algorithm,information entropy and memory-reduced algorithm to least square support vector machine(LS-SVM) on the basis of the traditional support vector machine prediction model and advances a memory-reduced type of prediction model of LS-SVM which is based on genetic algorithm and information entropy.The paper also independently derives the entropy fit for the financia...