| In the 21st century, there are a variety of opportunities for the listed corporations as the main body of the capital market, meanwhile they are faced with various challenges with the large and far reaching economy globalization. In 2008, the global financial crisis triggered by the U.S. subprime mortgage crisis, has issued a warning to the public. The bankruptcy of an enterprise not only the enterprise itself falls into a financial crisis, but as a food chain in the ecosystem there are a number of citizens and businesses will be involved in, even arouse a disaster. Simultaneously, by the end of 2009 the number of companies being special treatmented was accumulated up to 155 in Shanghai and Shenzhen stock exchanges, implying that a considerable number of the listed firms in China have got involved in financial distress. Admittedly, the company suffered from financial difficulties not only put operation pressure on their own business, but also investors, creditors and other related benefits are being affected. Therefore, concerning about listed company's financial status that will influence the company itself even the entensive capital market, is required to accurately distinguish healthy from unhealthy firms according to financial and non-financial differences. And listed companies can be more imformed with the aid of accurate prediction which is in a timely manner to prevent and defuse financial turmoil and run a healthy and sustainable business.Currently although there are a large number of predicting models for financial distress, problems cannot be negnected in the existing studies, mainly about the precision of the predicting results is not high, or its adoptable availability is unexpected. This paper therefore seeks to propose that adopting gray combination methods into the predicting model can selecte effective variables. Especially, the gray-logistic regression model is established to predict the probabilities of falling into financial distress for the corporations. The outlines are as follows.①The definition of financial distress is presented making use of the extant literature. And the theories of financial distress prediction are summarized utilizing some of the previous different perspectives. There is an overview focusing particularly on quantitative predicting models of corporate financial distress on the basis of the existing representative models.②The gray-logistic regression model is developed by the employment of the advantages of logistic model and the method of gray modeling, which is available to predict the failure probabilities when the sample size is smaller and some data of listed enterprises is incompelete or unexact.③The studied sample is consist of special treatmeted companies (ST companies) and the matched companies (non-ST companies), and the matching ratio is of 1:3, with a total sample of 120 (64 in 2009, 56 in 2010). The year that the company was special treamented will be defined as year t, using the 5 years data prior to the year t (t-1, t-2, t-3, t-4, t-5) as the sample to select effective variables, choosing the 4 years data prior to the year t (t-2, t-3, t-4, t-5) to build predicting models.④During the empirically testing process, the appropriate variables for the prediction model are collected by adopting the gray combination method, covering 9 categories of financial ratios (including 54 variables). There is a comparison in the predicting precision between the logistic regression model and gray-logistic regression model. According to the results, the accumulated maximum probability predicted correctly is around 85.9% and the maximum probability that companies will be special treatmented is around 62.5% tested by logistic regression model. However the accumulated minimum probability predicted correctly is around 89.3%, and the minimum probability that companies will be special treatmented is around 71.4% predicted by the advanced gray-logistic regression model, with its smallest goodness of fit test being around 0.899. |
PDF Full Text Request