Font Size: a A A

Research Of Financial Distress Prediction Model Based On Rough Sets And Support Yector Machines

Posted on:2012-11-18Degree:MasterType:Thesis
Country:ChinaCandidate:M WangFull Text:PDF
GTID:2249330395955677Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Financial distress is one of the serious problems that Chinese listed corporate faces. It is one of urgent tasks of financial management to establish an effective practical financial distress predicting model and to remove the ill influence of the financial crisis.In this paper, a hybrid prediction model based on rough set and support vector machines is proposed to predicting the financial distress of Chinese listed corporate in the first time. It has great significance of the investment and the government’s macro control.The main content of this research is organized as follows:Firstly, we review the achievements of financial distress prediction domestic and oversea in detail, define the concept of financial crisis on the basis of the empirical and drawback of other researches in this fields; we propose the indicator system of this study through discussing the advantages of cash-flow index.Secondly, we show the superiority of rough set theory by a lot of research. As the complexity of listed corporate financial distress prediction, the selection of indicators often relies on the experience of the experts. Rough set can be used for indicators selection to reduce the computation complexity and improve the efficiency of algorithms.Finally, we research the statistical learning theory and support vector machines systematically. And then a hybrid prediction model based on rough set and support vector machines, RSS model, is proposed to predicting listed corporate financial distress. The cash flow index is employed to test the proposed model. Firstly, RS is employed to reduce the financial index of listed corporate, then, the reduced index is input into the SVM for analysis, at the same time, a hybrid model-RSS model is constructed for listed corporate financial distress predicting. The results of experiments show that the application of RSS model is feasible and effective. RSS model can be used as a viable alternative solution for listed corporate financial distress prediction.
Keywords/Search Tags:rough sets, support vector machines, financial distress predictionartificial neural networks, cash flow
PDF Full Text Request
Related items