Font Size: a A A

Research On Earnings Quality Evaluation Of Zotye Automobile Co.,LTD.Based On BP Neural Network

Posted on:2024-05-09Degree:MasterType:Thesis
Country:ChinaCandidate:H D WangFull Text:PDF
GTID:2542306926972929Subject:Accounting
Abstract/Summary:
In today’s economic situation,the automotive industry,as an important aspect of the national economy,provides a strong impetus for social development.If China wants to change from a "big auto country" to a "strong auto country",we need to find out the problems in the development of the auto industry and solve them,and we must accurately predict the future development trend of the auto industry in order to promote the high-quality development of the auto industry.At present,the accounting system of China’s enterprises is still not sound,and the managers of enterprises are seeking their own selfish interests and thus undermining the quality of corporate surplus,or even falsifying financial information and providing false accounting information,which seriously disrupt the market order,deceive some stakeholders and damage their legitimate rights and interests.Therefore,the evaluation of surplus quality is of great concern to scholars and stakeholders in various countries,and how to accurately and reasonably evaluate surplus quality has become an urgent problem in the capital market.To address the above issues,this paper establishes a surplus quality evaluation index system based on surplus quality characteristics and comprehensive financial indicators for Zotye Auto from the financial perspective.Firstly,a scientific and reasonable index selection procedure is carried out using the Delphi method,and five primary indicators based on the characteristics of surplus quality and 18 secondary indicators are determined through two rounds of questionnaire and expert letter inquiry reliability analysis.Secondly,on this basis,the expected output value is determined by CRITIC assignment method,and the surplus quality evaluation model is established by using BP neural network,and the number of nodes of each layer of this three-layer BP neural network model is determined as 18-15-1,respectively.A total of 88 sets of data from 22 automotive OEMs during 2017-2020 are imported into the neural network model,and 1500 times of training and learning are conducted until The system is stopped when the error is reduced to an acceptable range,and the process of evaluating the surplus quality of automobile enterprises is completed.Then the constructed model was tested by simulation using 22 sets of data in 2021,and the results achieved the simulation accuracy requirements.Therefore,the constructed BP neural network model can effectively evaluate the surplus quality of automotive enterprises.Finally,an example analysis of Zotye Auto using the built model successfully verifies the validity and reasonableness of the BP neural network model and obtains the surplus quality score of Zotye Auto from 2017 to 2021.The evaluation results show that its surplus quality level is not satisfactory.Combining with the actual operation of the enterprise,this paper analyzes the reasons for the low surplus quality score of Zotye Auto,and proposes that Zotye Auto must focus on improving profitability,rational planning strategy,improving the enterprise asset management system,eliminating blind expansion,and increasing the core technology R&D and innovation,so as to obtain long-term sustainable development,which provides some reference for stakeholders.It provides some reference for stakeholders.
Keywords/Search Tags:Earnings quality characteristics, earnings quality evaluation, Back Propagation, factor analysis method
Related items