Font Size: a A A

Financial Early Warning For Property Insurance Company Based On Radial Basis Function Neural Network

Posted on:2011-12-20Degree:MasterType:Thesis
Country:ChinaCandidate:H WenFull Text:PDF
GTID:2219330371963688Subject:Finance
Abstract/Summary:PDF Full Text Request
Financial risk warning can be defined as such kind of a system, by obeying the relative insurance statutes and principles and choosing some indexes which can reflect the risk evidence of property insurance companies in advance, which establishes models and gives supervision and warning to financial risk in order to lead the supervision authorities and property insurance companies to pay attention to the deviation occurred in management, thus produce prevention in advance. It is the important process during the risk management of property insurance companies, and plays a vital role on the survival and development of them.Through the analysis of property insurance industry'actual situation, this paper, which based on economics, insurance theory and closely combined with the actual situation of our property insurance companies, do researches by comprehensively using neural network method and some basic principles such as financial risk management theory, finance, system theory, statistics and accounting, on the base of analyzing deeply the financial risk of property insurance companies. We take the mixed unbalanced panel data of 40 property insurance companies from 2006 to 2008 as the study samples and use radial basis function neural network model as the study method. Then we discuss the financial warning index system from three angles: solvency, profitability and growth capacity by using the complementary approaches of the analytical and empirical analysis and collecting data from the financial statements such as the balance sheet, profit flow table and cash flows. In addition, we overcome the subjective randomness of the subjective weighting method, since we use the variation coefficient method, which determines the weight of every index according to variability of the sample data, to determine the weight of various indicators for the property insurance company's financial index system.The empirical results indicates that this model can provide more effective information for the financial warning of the management,and it has bright prospects for the application under the situation that there are not enough samples and not many valid data because of the short history of insurance industry in China. And this paper, which does research on financial warning system of property insurance company, is in pursuit of providing theory guidance for companies those tend to establish financial warning system which includes tracking, monitoring and warning and discovering the problems exist in financial management in time so as to improve management and prevent failure.
Keywords/Search Tags:financial risk warning, mixed unbalanced panel data, variation coefficient method, radial basis function neural network
PDF Full Text Request
Related items