Research On Free Cash Flow Prediction Based On SVM In Power Generation Enterprise Valuation

In recent years,China’s power demand has been increasing.At the same time,the transfer,mergers and acquisitions and other transactions of power generation enterprises have become more frequent,and the evaluation demand has also increased.Among the income method to evaluate the value of power generation enterprises,the free cash flow discount method is mature and widely used.However,the traditional free cash flow forecasting method has some problems such as too subjective forecast of operating income and ignoring external factors.How to grasp the uncertainty of free cash flow forecasting and scientifically evaluate the value of power generation enterprises has become an urgent problem to be solved.Therefore,this paper aims to design a more reasonable free cash flow forecasting method through SVM,so as to make the evaluation result more reliable.Based on the literature review,this paper adopts the method of combining qualitative and quantitative analysis.Firstly,the concepts of free cash flow and SVM are introduced,and the methods of free cash flow discount method,support vector regression machine and singular spectrum analysis and prediction are expounded.Then,the influencing factors of power generation enterprise value and the limitations of existing traditional free cash flow forecasting methods in the application process are discussed.At the same time,the applicability of SVM to predict the free cash flow of power generation enterprises is further analyzed.Finally,the overall value of TP power plant is evaluated according to the scheme of SVM predicting free cash flow,and the change of the final prediction result accords with the historical trend,which verifies the advantages and scientificity of the improved prediction model compared with the traditional prediction model.Based on the literature review,this paper adopts the method of combining qualitative and quantitative analysis.Firstly,the concepts of free cash flow and SVM are introduced,and the methods of free cash flow discount method,support vector regression machine and singular spectrum analysis and prediction are expounded.Then,the influencing factors of power generation enterprise value and the limitations of existing traditional free cash flow forecasting methods in the application process are discussed.At the same time,the applicability of SVM to predict the free cash flow of power generation enterprises is further analyzed.Finally,the overall value of TP power plant is evaluated according to the scheme of SVM predicting free cash flow,and the change of the final prediction result accords with the historical trend,which verifies the advantages and scientificity of the improved prediction model compared with the traditional prediction model.This paper summarizes the research conclusions from two aspects.Firstly,it is concluded that the free cash flow of power generation enterprises is most sensitive to the change of operating income,and the operating income is mainly determined by factors such as average utilization hours of power generation equipment,power generation,electricity price and VAT rate.Secondly,it is concluded that SVM has good applicability and learning performance in forecasting the free cash flow of power generation enterprises,which overcomes the limitations of traditional forecasting models,makes the value of the evaluated power generation enterprises more scientific and accurate,and can be well applied to the evaluation practice.