基于PCA-SOM的上市公司综合绩效评价

为摆脱用传统方法评价上市公司财务状况综合绩效的局限，更好地为证券监管机构的监管和广大投资者的风险控制提供决策参考，提出了上市公司综合绩效评价的PCA-SOM（Principal Components Analysis-System Object Model）模型。该模型首先利用主成分分析（PCA：Principal Components Analysis）建立公司财务状况综合评价模型，然后利用自组织网络（SOM：System Object Model）模型对上述模型的有效性进行检验，并采用了2005年房地产业47家房地产上市公司的具体财务数据进行分析，最终给出了中国房地产上市公司财务水平的排名。结果表明，PCA-SOM模刑更好地模拟了实际数据，同时排除了主观因素的干扰，较为客观地评价了上市公司，给出了较为公止的排名，为监管机构提供了参考依据。

Financial Assessment of Listed Companies Based on PCA-SOM

In order to overcome the limitation of traditional method and provide a decision making reference for the supervision of securities organizations and the risk control of investors. A novel PCA-SOM （ Principal Components Analysis-System Object Model） is proposed for assess of financial performance of the listed companies. In this model, the PCA （ Principal Components Analysis） is utilized to establish an assessment model of a com- pany＇s financial performance. And the SOM （System Object Model） model is used to examine the effectiveness of the proposed model. The experiments are conducted on the financial dataset of 47 listed realty companies to examine the presented model. The simulated results show that the PCA-SOM model performs better, reduces the disturbance of manual factor effectively, appraises the listed companies, provides the currect ranking and offers the reference for the institution.