| 以下仅列出文章摘要、提纲简介，如需获取全文阅读权限，或原创定制、长期合作，请随时联系。 微信QQ:312050216 |
 |
 |
| 我国制造业上市公司财务风险评价及预警研究
中文摘要
制造业作为国民经济的主体，是中国经济的发展动力，是中国长期发展的关键所在，正因为制造业在中国经济发展中担任着至关重要的作用，财务风险管理是其不可忽视的问题。本文对国内外相关研究进行整理总结，在对制造业目前情况充分分析下，在制造业上市公司中选取2019年被ST的46家公司作为财务危机组，以及未被ST的46家公司作为财务健康组，共92家公司作为此次研究的样本公司。在样本公司中，进一步选取公司2016年－2018年24个财务指标数据，财务指标涉及偿债能力、营运能力、盈利能力、现金流量能力以及企业发展能力五个方面。通过对24个财务指标进行显著性检验，剔除掉不显著指标后，利用主成分分析的方法，对指标进行降维处理，最终得到6个主成分，总方差解释了原变量的77.6%。在主成分结果基础上，建立Logistic回归模型、LDA模型、朴素贝叶斯分类模型、SVM支持向量机模型、决策树模型以及随机森林模型六个财务预警模型，分析和比较六个模型结果后，发现随机森林的预测效果是最好的，总正确率为67.7%，其中1类正确率为70.8%，0类正确率为64.6%。在其他模型预测结果当中，朴素贝叶斯方法对于0类预测正确率达到89.6%，决策树方法对于1类预测正确率达到77.1%，所构建的预测模型对于制造业企业财务风险的预测有一定的效果。最后结合研究结果，提出了一些建议。本文对制造业上市公司财务风险评价和预警具有一定的参考作用。
关键词：制造业上市公司；财务风险；财务风险评价；财务风险预警；
ABSTRACT
As the main body of the national economy,manufacturing is the driving force of China's economic development and the key to China's long-term development. Precisely because manufacturing plays a vital role in Chinas economic development, financial risk management is a problem that can not be ignored. Based on the analysis of the current situation of the manufacturing industry, 46 companies listed in the manufacturing industry in 2019 were selected as the Financial Crisis Group,and 46 companies not listed by St as the Financial Health Group, a total of 92 companies as a sample of the study companies. In the sample companies, further select the company 2016-201824 financial indicators data, financial indicators related to solvency, operating capacity,profitability, cash flow capacity and enterprise development capacity in five areas. After testing the significance of 24 financial indicators,this paper puts forward the method of principal componentanalysis to reduce the dimension of the indicators after dropping the non-significant indicators. Based on the results of principal component analysis,six financial early warning models including Logisticregression model, LDA model,naive Bayes classification model, SVM Support vector machine model, decision tree model and random forest model are established, after analyzing and comparing the results of six models, it is found that the prediction effect of random forest is the best, the total correct rate is 67.7% , among which the correct rate of type 1 is 70.8% , and that of type 0 is 64.6% .Among other model prediction results, naive bayes method achieves 89.6% accuracy for class 0 prediction and decision tree method achieves 77.1% accuracy for class 1 prediction,the prediction model has certain effect on the prediction of financial risk of manufacturing enterprises. Finally, some suggestions are put forward based on the research results.This paper can be used as a reference for the financial risk assessment and early warning of listed manufacturing companies.
Keywords: listed manufacturing companies; financial risk;financial risk evaluation; financial risk early warning;
|
