A Predicting Study On Financial Crisis Of Listed Manufacturing Companies

After the accession of China to WTO, domestic companies are faced with much fiercer external competition, which has put forward higher requirements on internal management. Because of their own mismanagement, some listed companies have trapped into financial distress. The financial distress can not only be a threat to the companies’development and survival, but also inflict huge losses on creditors and investors. In order to survive and develop in this fierce competition, we have to build up financial early-warning system, against all risks of that the listed companies may come across, to predict and prevent the financial distress. The research on financial early-warning system in foreign countries is earlier than us, so they have accumulated lots of advanced management concepts and methods in risk management. If these ideas can be applied to our market economy, establishing a financial early-warning index system which is unrelated to each other but contains a large amount of information and a financial distress early-warning model with high prediction accuracy, they will play an important role in managing the risks in our listed companies of manufacturing industries.Firstly, according to related literature and basic theoretical knowledge of financial accounting, this paper shortlists18financial indices including current ratio, quick ratio, ratio of working capital to total assets, asset-liability rate, return on total assets, return on equity, net profit margin on sales, profit ratio on major business, inventory turnover ratio, accounts receivable turnover ratio, total capital turnover ratio, circulating capital turnover ratio, accounts receivable accounts for sales income ratio, ratio of cash to current debts, cash flow ratio, debt coverage ratio, ratio of cash sales and asset cash recovery rate. This paper regards ST and non-ST status of listed companies of manufacturing as a symbol to judge whether a company is in financial distress or not. With the principle of random sampling, the paper has selected30companies which are first announced into ST from2009to2011as samples of troubled companies and60companies which have never been in ST as the samples of normal companies. I build a financial early-warning model according to the financial indexes of the90listed companies of manufacturing and pick up their financial data from2002to2011. Secondly, the paper analyses the preliminary18indexes of the financial data by entropy method, and eventually determine10financial indexes:inventory turnover ratio, current ratio, quick ratio, ratio of cash sales, accounts receivable turnover ratio, accounts receivable turnover ratio, asset-liability rate, circulating capital turnover ratio, cash ratio, total capital turnover ratio---to be the influencing factors. Thirdly, this paper removes the co linearity in financial indexes and reduces the number of independent variables by principal component analysis, making the operation more convenient and easier. Finally, the paper builds the random effects regression model with panel logit by Housman test. The parameters are in line with the theoretical basis of the financial accounting, then I test the samples, and the correct rate of predicted result is81.67%.The basic framework of this paper:the fist part is introduction. This part mainly includes the introduction of this research’s background and significance. summary of empirical research on financial early-warning models at home and abroad, as well as the basic content and framework of this paper. The second part is theoretical research. This part introduces the related theories and early-warning principles of financial distress in manufacturing, in which the meaning and characteristics of financial distress, the basic concept and theories of financial early-warning system together with the analysis on manufacturing industry are mainly introduced. The third part is model design. This part presents a basic introduction to entropy method, principal component analysis and panel logit method, and studies the selection and samples of financial indexes. The forth part is empirical research. In this part, the financial indexes which are going to be employed in the model are decided with entropy method, and the financial early-warning model is built under the empirical research method that combined principal component analysis and panel logit method, at last the predictive tests and analysis are conducted. The fifth part is conclusion and suggestion, in which I evaluate the panel logit warning model and give recommendations for future research by analyzing the development situation of our domestic companies.The characteristics of this paper includes:first, after preliminary selection of financial indexes, with entropy method, I determine the final financial indexes which will be selected into the model. Most scholars, with strong subjectivity, select financial indexes based on the experience of other scholars’studies, which the selection of financial indexes by entropy method has certain theoretical basis and strong objectivity. Second, the model is panel logit model. The majority of research scholars make model and conduct prediction with cross-sectional data. Though the forecast is accurate, the choice of the cross-sectional data cannot reflect the dynamic facts of the companies’development. This paper selects10years’financial data of Chinese listed companies in manufacturing industries to make a research, what’s more, by adopting cross-sectional data and time series data, the capacity of samples has been greatly increased, with which the dynamic facts of companies’development can be reflect clearly. Moreover, forecast test shows that the model has great practicability.