Research On The Investment Value Of Equipment Manufacturing Enterprises Based On Machine Learning

As the basic industry to support the national economic development,the manufacturing industry is the main body of the national economy.To build a manufacturing industry with international competitiveness is the only way for China to improve the comprehensive national strength,safeguard the national security and build a world power.The equipment manufacturing industry is the basic and strategic industry for the development of the national economy,which reflects the scientific and technological level,the manufacturing capacity and the comprehensive strength of a country or region.If we want to enhance our comprehensive national strength and vigorously develop manufacturing industry,we must think about equipment manufacturing industry.Advanced equipment manufacturing,led by high and new technology,is the core of equipment manufacturing industry,the backbone of modern equipment manufacturing industry,and the engine of industrial transformation and upgrading.It is of great significance to accelerate the cultivation of new economic growth points,promote industrial structure adjustment and build strong industrial province.National policies have been encouraging and promoting the innovation and development of manufacturing and equipment manufacturing industry.GEM listed companies are mostly in the stage of rapid growth,with the characteristics of high technology and innovation,but at the same time,they are accompanied by high risk.Under the encouragement of policy,a large number of scientific and technological manufacturing enterprises appear and develop rapidly.Under the operation of capital,some enterprises will get the favor of investors.It seems that how to select enterprises with promising prospects is an investor's concern.From the perspective of investors,this paper considers five aspects of profitability,growth ability,solvency,operating ability,and governance ability,and uses the method of combining the random forest and the correlation coefficient to evaluate and select the importance of the 142 financial indicators,and finally sifting the long-term debt ratio,the liquidity ratio,and the financial leverage.Operating lever,capital accumulation rate,net profit growth rate,growth rate of operating income,capital density,total assets turnover,receivable turnover,fixed assets and income ratio,operating gross profit rate,net asset return rate,net profit rate of fixed assets,profit rate of cost and cost,management cost rate,A stock number,The final index system is the 20 high importance and no direct correlation index of the first big shareholders,the top three directors' total pay and the proportion of top executives.Support vector machine is applied to build a classification evaluation model of investment value of equipment manufacturing enterprises.By using 20 index data in five aspects of the previous year as the input data of the classification model,the comparison result of the monthly average rate of return and the monthly average return rate of the enterprise for second years is used as the output end classification standard: if the monthly average yield of the enterprise is greater than zero,it is larger than the average return of the gem.The enterprise is worth investing,and it is recorded as 1-YES.Otherwise,the enterprise is not worth investing,it is recorded as 0-NO.The sample input data is the 20 financial index data of 2012-2016 years in the equipment manufacturing industry of the national Tai'an database gem.The output is compared with the average return of 2013-2017 years and the gem,and the data of the two ends are automatically matched for 1 years.The sample data are preprocessed,then they are randomly divided into training set and test set,and the training and testing of the model is completed.The parameters gamma and nu are adjusted and the optimal model is obtained.And the model is used to classify the investment value forecast of the 2017 enterprises which have been published in the annual financial report.Among the 5 enterprises that publish the complete data,2 are worth 300161 and 300590,respectively,300102 and 300545 are worth investing in the 7 enterprises with the missing value of the index set;The over analysis proves that these four enterprises have high investment value,which proves that the classification model has high accuracy for the prediction and evaluation of the investment value of the equipment manufacturing enterprises on the gem.It can provide reference for investors.