Font Size: a A A

The Study On The Method Of Business Performance Evaluation Of Listed Companies In China And Its Application

Posted on:2008-03-14Degree:MasterType:Thesis
Country:ChinaCandidate:H S DingFull Text:PDF
GTID:2189360215950910Subject:Management Science and Engineering
Abstract/Summary:PDF Full Text Request
With the gradual normalization of our national stock market and the great expansion of the number of the listed companies, the shareholders, managers and authorities of government show great concern on the performance of companies in China at present. The proper, just, and reasonable performance evaluation of listed companies and the horizontal and vertical comparison between these companies's performances make the related departments and authorities know the developing situation of company and what ranks and positions they are in the same industry, which could supervise and stimulate these managers in these companies and give some references to the investors.The financial indices of listed companies are more than 100 terms and very complex, moreover, common terms used frequently amount to tens. Due to the limited ability and energy of the average shareholders, it is very unrealistic for them to gain dozens and even hundreds of financial indices and evaluate. In this paper, the method integrating K-clustering with principal component analysis to decide the classification and indices of listed companies is proposed and comprehensive evaluation model with multivariate time series objective weighting based on difference driving is built. In this empirical test, business performance evaluation of listed companies in vehicle trade is chosen as object to study. Considering the continuity and integrality of financial data, this paper selects 28 stocks in vehicle trade as well as 32 financial indices such as earning per share of stock, net asset per stock, yield ratio of net asset of three successive years of 2003-2005, etc. At the same time, using K-means clustering, the number of cluster can be determined according to the size of the F-means value, and then the indices of classes including more indices are analyzed by the method of principal component analysis. Considering the real conditions together with actual meaning of indices, 12 indices such as earnings per share of stock, net asset per stock, yield ratio of net asset, accumulation fund of asset per stock, cash flux of management per stock , net account receivable , net merchandise inventory , long term net investment, total assets, total liabilities, net profit, fixed assets depreciation and so on are chosen from 32 indices at last. By the comprehensive evaluation model with multivariate time series objective weighting based on difference driving, the comprehensive evaluating values are gained. Comparing the evaluation results through 12 indices and 32 indices with the ranks from the website www.stockstar.cm, the scientific validity of the method discussed in this paper is verified.
Keywords/Search Tags:listed companies, K-means clustering, F-statistic, principal component analysis, Multivariate time series evaluation
PDF Full Text Request
Related items