节点文献

基于BP神经网络模型的鹏都农牧财务风险预警研究

Research on Early Warning of Financial Risks of Agriculture and Animal Husbandry in Pengdu Based on BP Neural Network Model

【作者】 李琳

【导师】 张志;

【作者基本信息】 山东农业大学 , 会计(专业学位), 2023, 硕士

【摘要】 随着我国经济的飞速发展,农业第一产业的地位也愈加凸显,农业对于国民经济的稳定发展起着至关重要的作用。且与其他行业企业相比,农业存在其固有的特定风险,因此构建一套科学有效的财务风险预警模型,提升农业上市公司财务风险控制能力具有重要意义。鹏都农牧作为一家综合性农业产业化企业,存在着内外部多种问题,具体情况如下:鹏都农牧虽并未戴上“ST”的危险帽子,甚至观其财报企业自上市后基本每年的净利润都是正数,表面上实现了盈利。但实际上该企业上市九年以来的扣非净利润均为负数,2021年六月份,鹏都农牧因此被证监会问询,非经常性损益由于与企业发生的经营业务无直接关系,且难有连续性,会影响企业正常盈利能力真实、公允的反映。而鹏都农牧的扣非净利润多年连续为负,一定程度上反映了其并未实现实际意义上的盈利或盈利能力匮乏。除此之外,近年来鹏都农牧可以说是深交所的“重点关注对象”,年报问询函是家常便饭。加之诸如主营业务频频变换、大股东高股权质押率、公司有息负债规模超过货币资金规模等其他因素,故认为鹏都农牧需要进行财务风险预警,控制风险。本文以在A股上市的35家农业上市公司2015-2020年24个季度的财务数据为样本,选取了29个财务指标和1个非财务指标,使用因子分析法提取了10个主因子,通过BP神经网络模型的训练和检验,构建财务风险预警模型,对鹏都农牧进行财务风险预警研究。模型经过训练和检验后,整体的预测准确率达到92.1%,对鹏都农牧进行模型应用,发现预警结果和实际情况较为相符。即选用BP神经网络,构建财务风险预警模型,进行风险预警控制,有利于得到更加客观精准的数据结果,能够为鹏都农牧进行财务风险预警工作提供参考。

【Abstract】 With the rapid development of China’s economy,the status of the primary industry of agriculture has become more prominent,and agriculture plays a vital role in the stable development of the national economy.Compared with enterprises in other industries,agriculture has its inherent specific risks,so it is of great significance to build ascientific and effective financial risk early warning model to improve the financial risk control ability of agricultural listed companies.As a comprehensive agricultural industrialization enterprise,Pengdu Agriculture and Animal Husbandry has a variety of internal and external problems,the specific situation is as follows: Although Pengdu Agriculture and Animal Husbandry does not wear the dangerous hat of "ST",and even looks at its financial reports,the net profit of its financial reporting enterprises has basically been positive every year since its listing,and it has achieved profits on the surface.In June 2021,Pengdu Agriculture and Animal Husbandry was inquired by the China Securities Regulatory Commission,and the non-recurring profit and loss is not directly related to the business business of the enterprise and is difficult to have continuity,which will affect the true and fair reflection of the normal profitability of the enterprise;Therefore,we believe that this indicator should be excluded when assessing enterprise value.The non-net profit of Pengdu Agriculture and Animal Husbandry has been negative for many years,to a certain extent,reflecting that it has not achieved profitability or lack of profitability in the actual sense.In addition,in recent years,Pengdu Agriculture and Animal Husbandry can be said to be the "focus of attention" of the Shenzhen Stock Exchange,and the annual report inquiry letter is commonplace.In addition,other factors such as frequent changes in the main business,high equity pledge ratio of major shareholders,and the scale of the company’s interest-bearing liabilities exceeding the scale of monetary funds,it is believed that Pengdu Agriculture and Animal Husbandry needs to warn of financial risks and control risks.This paper takes the financial data of 35 agricultural listed companies listed on A-shares in 24 quarters from 2015 to 2020 as samples,selects 29 financial indicators and 1non-financial index,extracts 10 principal factors by factor analysis method,constructs a financial risk early warning model through the training and testing of BP neural network model,and conducts financial risk early warning research on Pengdu agriculture and animal husbandry.After training and testing,the overall prediction accuracy of the model reached92.1%,and the model application was carried out on Pengdu Agriculture and Animal Husbandry,and it was found that the early warning results were more consistent with the actual situation.That is,the BP neural network is selected to build a financial risk early warning model and carry out risk early warning control,which is conducive to obtaining more objective and accurate data results,and can provide reference for Pengdu Agriculture and Animal Husbandry to carry out financial risk early warning work.

  • 【分类号】F324;F302.6
节点文献中: 

本文链接的文献网络图示:

本文的引文网络