[1]肖叶宇,张 闪.基于随机生存森林的企业财务危机研究[J].南京师大学报(自然科学版),2021,44(04):1-6.[doi:10.3969/j.issn.1001-4616.2021.04.001]
 Xiao Yeyu,Zhang Shan.Research on Financial Crisis of Enterprises Based on Random Survival Forest[J].Journal of Nanjing Normal University(Natural Science Edition),2021,44(04):1-6.[doi:10.3969/j.issn.1001-4616.2021.04.001]
点击复制

基于随机生存森林的企业财务危机研究()
分享到:

《南京师大学报(自然科学版)》[ISSN:1001-4616/CN:32-1239/N]

卷:
第44卷
期数:
2021年04期
页码:
1-6
栏目:
·数学·
出版日期:
2021-12-15

文章信息/Info

Title:
Research on Financial Crisis of Enterprises Based on Random Survival Forest
文章编号:
1001-4616(2021)04-0001-06
作者:
肖叶宇1张 闪2
(1.吉林大学数学学院,吉林 长春 130012)(2.南京财经大学应用数学学院,江苏 南京 210023)
Author(s):
Xiao Yeyu1Zhang Shan2
(1.College of Mathematics,Jilin University,Changchun 130012,China)(2.School of Applied Mathematics,Nanjing University of Finance and Economics,Nanjing 210023,China)
关键词:
随机生存森林生存分析财务危机预警
Keywords:
Random Survival Forestsurvival analysisfinancial crisis early warning
分类号:
O213
DOI:
10.3969/j.issn.1001-4616.2021.04.001
文献标志码:
A
摘要:
以沪深两市A股制造业上市公司为样本,将随机生存森林模型引入企业财务危机研究中去. 通过计算两种度量下变量重要性排名,发现营业收入增长率和息税前利润对财务危机的影响最大. 随后将随机生存森林与Cox、后向逐步Cox和Lasso-Cox模型进行对比,随机生存森林的预测性能要优于3种Cox模型. 同时结合随机生存森林下的生存函数和累积风险函数,对公司被特别处理的时间进行分析,结果显示模型有很好的预警功效,可以为各利益相关方的决策提供依据.
Abstract:
Taking Shanghai and Shenzhen A-share manufacturing listed companies as a sample,the Random Survival Forest model is introduced into the research on corporate financial crisis. It is found that the growth rate of operating income and the profit before interest and tax have the greatest impact on the financial crisis by calculating the ranking of the importance of variables under the two measures. Subsequently,the Random Survival Forest was compared with Cox model,Cox model with backward stepwise variable selection and Lasso-Cox models. The prediction performance of the Random Survival Forest is better than the three Cox models. At the same time,combined with the survival function and cumulative hazard function under the random survival forest,the company is analyzed for the time when the company is ST. The results show that the model has a good early warning function,which can provide a basis for the decision-making of various stakeholders.

参考文献/References:

[1] 李扬,李竟翔,马双鸽. 不平衡数据的企业财务预警模型研究[J]. 数理统计与管理,2016,35(5):893-906.
[2]马超群,何文. 基于Cox的财务困境时点预测模型研究[J]. 统计与决策,2010(21):38-42.
[3]王小燕,袁欣. 基于惩罚组变量选择的Cox财务危机预警模型[J]. 系统工程,2018,36(3):113-121.
[4]KALBFLEISCH J D,PRENTICE R L. The statistical analysis of failure time data[M]. 2nd ed. New Jersey:John Wiley & Sons,Inc,2002.
[5]SUN J. The statistical analysis of interval-censored failure time data[M]. New York:Springer,2006.
[6]高珍,柯阿香,余荣杰,等. 基于随机生存森林的交通事件持续时间预测[J]. 同济大学学报(自然科学版),2017,45(9):1304-1310.
[7]ISHWARAN H,KOGALUR U B,BLACKSTONE E H,et al. Random survival forests[J]. The annals of applied statistics,2008,2(3):841-860.
[8]王呈斌,方匡南,郑陈璐. 基于随机生存森林的房屋贷款逾期研究[J]. 上海金融,2020(2):59-63.
[9]MOGENSEN U B,ISHWARAN H,GERDS T A. uating random forests for survival analysis using prediction error curves[J]. Journal of statistical software,2012,50(11):1-23.
[10]KIM Y,PARK S,LEE J. Integrated survival model for predicting patent litigation hazard[J]. Sustainability,2021,13(4):1763.
[11]ISHWARAN H,KOGALUR U B,GORODESKI E Z,et al. High-dimensional variable selection for survival data[J]. Journal of the American statistical association,2010,105(489):205-217.
[12]鲍新中,陶秋燕,傅宏宇. 基于变量聚类和Cox比例风险模型的企业财务预警研究[J]. 系统管理学报,2015,24(4):517-523,529.

相似文献/References:

[1]张大鹏,程学亮,孙明霞.DeephitTM:医学生存分析的时间相关性深度学习模型[J].南京师大学报(自然科学版),2024,(03):138.[doi:10.3969/j.issn.1001-4616.2024.03.017]
 Zhang Dapeng,Cheng Xueliang,Sun Mingxia.DeephitTM:a Time-dependent Deep Learning Model for Medical Survival Analysis[J].Journal of Nanjing Normal University(Natural Science Edition),2024,(04):138.[doi:10.3969/j.issn.1001-4616.2024.03.017]

备注/Memo

备注/Memo:
收稿日期:2021-03-12.
基金项目:国家自然科学基金项目(11601224).
通讯作者:张闪,博士,副教授,研究方向:生物数学. E-mail:shanzhang86@163.com
更新日期/Last Update: 2021-12-15