基于主成分分析和归一化RBF神经网络优化的财务综合评价

The Comprehensive Evaluation of Finance Based on Principal Component Analysis and Normalized RBF Neural Networks Optimization

提出基于主成分分析和归一化RBF神经网络优化相结合的财务综合评价方法,克服传统神经网络财务分析中的评价指标偏少、存在局部极小等不足,为公司财务评价提供新的思路和方法,并利用神经网络模型对我国钢铁业上市公司2007年财务状况进行仿真实验,为分析者决策提供准确、可靠的参考依据。

This paper puts forward the comprehensive appraisal method based on principal component analysis and neural network optimization of normalization RBF, which overcomes the limitations of financial analysis in traditional neural network model, such as insufficient evaluation indexes and local minimization. It offers a new thinking and method for company financial appraisal. The model of neural network is applied to the simulation experiment on analyzing the financial situation of listed companies of steel industry in 2007 in our country and offers accurate and reliable reference basis for practical policy-making.