Research On The Design And Application Of The Discount Rate In Business Valuation Based On BP Neural Network

Along with the deepening of the state-owned assets management system, economic activities of non-state-owned enterprises have become increasingly frequent, the business valuation has become increasingly important part of market economy.It plays an important role in merger, investment and management. Income approach as one of the basic methods of asset valuation, is also the most widely used method about business valuation.which results is the present price of the company’s future earnings.The discount rate is an important parameter of the method, which has great influence on the final results of the assessment, so the use of scientific and reasonable method for determining the discount rate in the valuation of companies is particularly important. In view of this, the article is to do some explore and research on the method for determining the discount rate for business valuation.In this paper, I attempting to determine the discount rate by thinking BP neural network.BP neural network is a multi-level former feed forward neural network, which has strong nonlinear mapping ability.BP neural network has characteristics of high-speed computing and learning,and has the ability to approximate any nonlinear continuous function. Under conditions of inadequate information, BP neural network in forecasting, evaluation and other aspects of the application can achieve better results. BP neural network can overcome some shortcomings of current equity valuation methods such as lack of information, subjective in valuation and so on.The valuation result of BP neural network are more objectively and accurately. The principles of discount rate determine based on BP neural network are:choosing a number of companies which in the same industry and getting the relevant data of listed companies in as the input data to establish the discount rate simulation model. Through Matlab simulation program to train the study samples, then forming a general model which is used to simulated the discount rate of evaluated enterprise. And taking examples to compare this method with other methods, such as WACC, CAPM and so on. By comparing the value of the assessed value and the actual transaction value of the enterprise BP neural network to verify the applicability of the method. It’s showed that the discount rate is determined by the BP neural network, and then estimate the enterprise value is closer to the actual transaction price, BP neural network can make up for some shortcomings, and improve the equity valuation accuracy to some extent.BP neural network has some flaws,but the discount rate determine method based on BP neural network shares a new ideal and provides assistance for improving discount rate theory.