Font Size: a A A

Research On Financial Risk Early Warning Of Listed Companies On GEM

Posted on:2019-01-06Degree:MasterType:Thesis
Country:ChinaCandidate:C WangFull Text:PDF
GTID:2429330545957639Subject:Accounting
Abstract/Summary:PDF Full Text Request
Since second-board market was listed in 2009,it has attracted the attention of all walks of life and the favor of the investors in the 10 years or so.In China,its status in the market is just inferior to main-board market.But as time flies,the high profitability in the beginning and high growth rate began to weaken.At the same time,with the industrial adjustment and increasing market risk,as well as its features of unstable development,small scale,weak risk management consciousness,unsound rules and regulations,the financial risks are increasingly more prominent.For the second-board market enterprises,it is necessary to establish an appropriate financial risk prevention system as soon as possible so as to cope with the possible financial difficulties and it is beneficial to improve the enterprise risk management system and help the enterprise to prevent the risk as early as possible,having great significance for the healthy and sustainable development of enterprises in a rapidly changing market environment.Meanwhile,it has important reference value and function in effectively guiding the development of the second-board market,protecting the interests of investors and creditors and promoting the effective supervision of the relevant financial institutions.This paper summarizes the research results and early warning methods of the theory of enterprise financial risk of the experts and scholars at home and abroad with the macro market conditions and the financial situation and features of second-board market manufacturing enterprises in recent years taken into account,first selecting 43 pairs of sample data of financial health and financial crisis paired enterprises according to the annual report data of the second-board market manufacturing enterprises in 2016 and selecting 25 indicators from 5 dimensions of debt paying ability,profitability,development capability,operation capability and capital structure according to the public information on the NetEase Finance and the database of Genius Finance and conducting a comparative analysis between the ratios of financial risk in the first three years: T-1,T-2 and T-3 of the sample companies and the ratios of normal operating companies.The BP neural network method is chosen as the main research method of this paper,and the program is written in the Matlab language to reduce the dimension of the primary index using PCA principal component analysis,and the 17 financial ratios processed are taken as the input layer neurons of neural network in this paper;after continuous training and learning and considering the training error and the number of iterations,this paper considers that when the numbers of neurons in the hidden layer are both 5,the error and accuracy are the best,and a three-level neural network model with two hidden layers and one output layer is thus established.Simulation processing is conducted on the network model using test samples and it can be seen that the BP neural network model set up in this paper has an early warning accuracy of 88% after program operation.In order to verify the scientificalness and accuracy of the early warning model established in this paper,the prediction results of the multivariate variable F fractional model and conditional probability model are calculated,and the early warning accuracy of the three models are compared,which contribute to the conclusion that the BP neural network has the best warning effect and the highest accuracy rate.Finally,the stylistic entertainment and Internet enterprises of the second-board market are used to conduct second examination of the models,of which the accuracy rate is up to 80%.The results show that the BP neural network is suitable for early warning of financial risks in China's second-board market.
Keywords/Search Tags:Financial Alert, GEM, BP Neural Network
PDF Full Text Request
Related items