基于小波支持向量机的企业财务长期预测算法  

Long-term Prediction Algorithm of Enterprise Finance Based on Wavelet Support Vector Machine

作　　者：李珊珊 张福泉 LI Shanshan;ZHANG Fuquan(Department of Accounting,Fujian Business University,Fuzhou 350012,China;School of Computer Science and Technology,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]福建商学院会计系,福州350012 [2]北京理工大学计算机学院,北京100081

出　　处：《长春大学学报》2020年第1期31-35,共5页Journal of Changchun University

基　　金：福建省科技厅项目(2018H0028)

摘　　要：为了提高企业财务长期预测算法的准确度,提出采用基于小波支持向量机的策略来完成财务预测。将基本小波变换与神经网络相结合,完成财务预测指标特征向量的提取,借助支持向量机的方法对特征向量进行有效降维并得到最优解,再通过最优解来完成分类,从而预测企业财务运行情况。通过将预测结果与样本实际数据对比,证明提出的算法预测准确度高,在企业财务长期预测方面有较强的适用性。In order to improve the accuracy of long-term financial prediction algorithm,a strategy based on wavelet support vector machine is given to complete financial prediction. The combination of basic wavelet transform and neural network is used to extract the feature vectors of financial prediction indicators,the effective dimension reduction on the feature vectors is made by support vector machine to obtain the optimal solution,and the classification is completed through the optimal solution,so as to predict the financial operation of enterprises. Experiments show that the prediction accuracy of this algorithm is high by comparing the prediction results with the actual data of samples,having a strong applicability in the long-term financial prediction of enterprises.

关 键 词：小波变换 企业财务长期预测 神经网络 支持向量机 

分 类 号：F275[经济管理—企业管理]