基于Aalen可加模型的中国上市公司ST预测
|
摘要
针对上市公司特别处理(简称ST)政策,基于公司连续2年净利润为负的特征,利用生存分析方法,建立Aalen可加模型预测上市公司财务困境。采用中国制造业公司自上市日起到2015年的财务数据进行实证分析,讨论上市公司违约概率与财务预警指标间的关系。研究结果表明,总资产规模、营业利润率、运营资金/资产总金额以及留存收益/资产总金额4个指标均影响上市公司陷入财务困境的强度;除了总资产规模对其影响是常数外,其他3个指标对其影响均具有时变性;且总资产规模越大,该公司陷入财务困境的可能性越小。
Based on the characteristics of the negative net profit in the company for two consecutive years in view of the policy on the special treatment(ST), in this paper, an Aalen add-on model was established to predict the financial distress of the listed companies using the survival analysis method. An empirical analysis was conducted to discuss the relationship between the default probability of listed companies and financial early warning indicators using the financial data of the China's manufacturing companies from it's listing date up to 2015. The results show that regardless of the number of samples, the four important factors impacting the intensity of listing companies in financial distress are total asset size, operating profit rate, operating capital/total assets and retained earnings/total assets. In addition to the fact that the total asset size is constant, the other three indicators have the time-varying impact. And the larger the asset size, the less possibility the company is in financial distress.
Based on the characteristics of the negative net profit in the company for two consecutive years in view of the policy on the special treatment(ST), in this paper, an Aalen add-on model was established to predict the financial distress of the listed companies using the survival analysis method. An empirical analysis was conducted to discuss the relationship between the default probability of listed companies and financial early warning indicators using the financial data of the China's manufacturing companies from it's listing date up to 2015. The results show that regardless of the number of samples, the four important factors impacting the intensity of listing companies in financial distress are total asset size, operating profit rate, operating capital/total assets and retained earnings/total assets. In addition to the fact that the total asset size is constant, the other three indicators have the time-varying impact. And the larger the asset size, the less possibility the company is in financial distress.
引文
[1] Altman E I. Financial ratios, discriminant analysis and the prediction of corporate bankruptcy [J]. Journal of Finance, 1968, 23(4): 589-609.
[2] Pam B W. Discriminant analysis and the prediction of corporate bankruptcy in the banking sector of nigeria[J]. International Journal of Finance & Accounting, 2013, 2(6): 319-325.
[3] 运迪, 周建辉. 基于改进Z值模型的企业信用风险评估与检验[J]. 统计与决策, 2014(10): 173-176.
[4] Ohlson J A. Financial ratios and the probabilistic prediction of bankruptcy[J]. Journal of Accounting Research, 1981, 18(1): 109-131.
[5] Shumway T. Forecasting bankruptcy more accurately: A simple hazard model[J]. Journal of Business, 1999, 74(1): 101-124.
[6] 王宗军, 李红侠. 上市公司治理因素与财务困境关系的实证研究[J]. 管理学报, 2007, 4(6): 803-814.
[7] 章之旺.预期财务困境成本在债务违约预测模型中的信息含量[J].数量经济技术经济研究, 2008(3): 149-159.
[8] 邓晶, 秦涛, 黄珊. 基于Logistic模型的我国上市公司信用风险预警研究[J]. 金融理论与实践, 2013(2): 22-26.
[9] 谢赤, 赵亦军, 李为章. 基于CFaR模型与Logistic回归的财务困境预警研究[J]. 财经理论与实践, 2014, 35(1): 57-62.
[10] 刘京军, 秦宛顺.上市公司陷入财务困境可能性研究[J]. 金融研究, 2006(11): 44-52.
[11] 鲍新中, 陶秋燕, 傅宏宇. 基于面板离散选择模型的企业财务预警及其行业适用性[J].系统管理学报, 2016, 25(1): 36-44.
[12] Henebry K L. Do cash flow variables improve the predictive accuracy of a Cox proportional hazards model for bank failure? [J]. Quarterly Review of Economics & Finance, 1996, 36(3): 395-409.
[13] Lando D, Sk?deberg T M. Analyzing rating transitions and rating drift with continuous observations[J]. Journal of Banking and Finance, 2002, 26(23): 423-444.
[14] Fledelius P, Lando D, Perch N J. Non-parametric analysis of rating transition and default data[J]. Journal of Investment Management, 2004, 2(2): 77-100.
[15] Das S R, Darrell D, Nikunj K, et al. Common failings: How corporate defaults are correlated[J]. Journal of Finance, 2007, 62(1): 93-117.
[16] Duffie D, Saita L, Wang K. Multi-period corporate default prediction with stochastic covariates[J]. Journal of Financial Economics, 2007, 83(3): 635-665.
[17] Duffie D, Eckner A, Guillaume H, et al. Frailty correlated default[J]. Journal of Finance, 2009, 64(5): 2089-2123.
[18] Lando D, Nielsen M S. Correlation in corporate defaults: Contagion or conditional independence? [J]. Journal of Financial Intermediation 2010, 19(3): 355-372.
[19] Sudheer C, Catalina S, Stuart T. Modeling the loss distribution[J]. Management Science, 2011, 57(7): 1267-1287.
[20] Figlewski S, Frydman H, Liang Weijian. Modeling the effect of macroeconomic factors on corporate default and credit rating transitions[J]. International Review of Economics & Finance, 2012, 21(1): 87-105.
[21] 倪中新, 张杨. 基于Cox比例危险模型的制造业财务困境恢复研究[J]. 统计与信息论坛, 2012, 27(1): 15-20.
[22] 朱永明, 邵庚云.基于Cox生存模型的上市公司财务困境预测研究[J]. 商业研究, 2013, 55(11): 108-116.
[23] 田军, 周勇. 信用传染违约Aalen加性风险模型[J]. 应用数学学报, 2012, 35(3): 408-420.
[24] Lando D, Medhat M, Nielsen M S, et al. Additive intensity regression models in corporate default analysis [J]. Journal of Financial Econometrics, 2013, 11(3): 443-485.
[25] 杨淑娥, 王乐平. 基于BP神经网络和面板数据的上市公司财务危机预警[J]. 系统工程理论与实践, 2007, 27(2): 61-67.
[26] 谢赤, 罗长青, 王蓓. 财务困境预警SVM模型的构建及实证研究[J]. 当代经济科学, 2007, 29(6): 96-100.
[27] 鲍新中. 基于粒子群的K均值算法和粗糙集理论的财务预警[J]. 系统管理学报, 2012, 21(4): 461-469.
[28] 倪志伟, 薛永坚, 倪丽萍,等. 基于流形学习的多核SVM财务预警方法研究[J]. 系统工程理论与实践, 2014, 34(10): 2666-2674.
[29] 蔡玉兰, 崔毅. Merton违约距离模型对企业财务困境的预测能力研究——基于离散时间风险模型的实证分析[J]. 预测, 2015, 34(6): 33-38.
[30] 鲍新中, 杨宜. 基于聚类-粗糙集-神经网络的企业财务危机预警[J]. 系统管理学报, 2013, 22(3): 358-365.
[31] Mckeague I W, Sasieni P D. A partly parametric additive risk model[J]. Biometrika, 1994, 81(3): 501-514.
[32] Martinussen T, Scheike T H. Dynamic regression models for survival data[M]. New York: Springer, 2006.
[33] 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.
[2] Pam B W. Discriminant analysis and the prediction of corporate bankruptcy in the banking sector of nigeria[J]. International Journal of Finance & Accounting, 2013, 2(6): 319-325.
[3] 运迪, 周建辉. 基于改进Z值模型的企业信用风险评估与检验[J]. 统计与决策, 2014(10): 173-176.
[4] Ohlson J A. Financial ratios and the probabilistic prediction of bankruptcy[J]. Journal of Accounting Research, 1981, 18(1): 109-131.
[5] Shumway T. Forecasting bankruptcy more accurately: A simple hazard model[J]. Journal of Business, 1999, 74(1): 101-124.
[6] 王宗军, 李红侠. 上市公司治理因素与财务困境关系的实证研究[J]. 管理学报, 2007, 4(6): 803-814.
[7] 章之旺.预期财务困境成本在债务违约预测模型中的信息含量[J].数量经济技术经济研究, 2008(3): 149-159.
[8] 邓晶, 秦涛, 黄珊. 基于Logistic模型的我国上市公司信用风险预警研究[J]. 金融理论与实践, 2013(2): 22-26.
[9] 谢赤, 赵亦军, 李为章. 基于CFaR模型与Logistic回归的财务困境预警研究[J]. 财经理论与实践, 2014, 35(1): 57-62.
[10] 刘京军, 秦宛顺.上市公司陷入财务困境可能性研究[J]. 金融研究, 2006(11): 44-52.
[11] 鲍新中, 陶秋燕, 傅宏宇. 基于面板离散选择模型的企业财务预警及其行业适用性[J].系统管理学报, 2016, 25(1): 36-44.
[12] Henebry K L. Do cash flow variables improve the predictive accuracy of a Cox proportional hazards model for bank failure? [J]. Quarterly Review of Economics & Finance, 1996, 36(3): 395-409.
[13] Lando D, Sk?deberg T M. Analyzing rating transitions and rating drift with continuous observations[J]. Journal of Banking and Finance, 2002, 26(23): 423-444.
[14] Fledelius P, Lando D, Perch N J. Non-parametric analysis of rating transition and default data[J]. Journal of Investment Management, 2004, 2(2): 77-100.
[15] Das S R, Darrell D, Nikunj K, et al. Common failings: How corporate defaults are correlated[J]. Journal of Finance, 2007, 62(1): 93-117.
[16] Duffie D, Saita L, Wang K. Multi-period corporate default prediction with stochastic covariates[J]. Journal of Financial Economics, 2007, 83(3): 635-665.
[17] Duffie D, Eckner A, Guillaume H, et al. Frailty correlated default[J]. Journal of Finance, 2009, 64(5): 2089-2123.
[18] Lando D, Nielsen M S. Correlation in corporate defaults: Contagion or conditional independence? [J]. Journal of Financial Intermediation 2010, 19(3): 355-372.
[19] Sudheer C, Catalina S, Stuart T. Modeling the loss distribution[J]. Management Science, 2011, 57(7): 1267-1287.
[20] Figlewski S, Frydman H, Liang Weijian. Modeling the effect of macroeconomic factors on corporate default and credit rating transitions[J]. International Review of Economics & Finance, 2012, 21(1): 87-105.
[21] 倪中新, 张杨. 基于Cox比例危险模型的制造业财务困境恢复研究[J]. 统计与信息论坛, 2012, 27(1): 15-20.
[22] 朱永明, 邵庚云.基于Cox生存模型的上市公司财务困境预测研究[J]. 商业研究, 2013, 55(11): 108-116.
[23] 田军, 周勇. 信用传染违约Aalen加性风险模型[J]. 应用数学学报, 2012, 35(3): 408-420.
[24] Lando D, Medhat M, Nielsen M S, et al. Additive intensity regression models in corporate default analysis [J]. Journal of Financial Econometrics, 2013, 11(3): 443-485.
[25] 杨淑娥, 王乐平. 基于BP神经网络和面板数据的上市公司财务危机预警[J]. 系统工程理论与实践, 2007, 27(2): 61-67.
[26] 谢赤, 罗长青, 王蓓. 财务困境预警SVM模型的构建及实证研究[J]. 当代经济科学, 2007, 29(6): 96-100.
[27] 鲍新中. 基于粒子群的K均值算法和粗糙集理论的财务预警[J]. 系统管理学报, 2012, 21(4): 461-469.
[28] 倪志伟, 薛永坚, 倪丽萍,等. 基于流形学习的多核SVM财务预警方法研究[J]. 系统工程理论与实践, 2014, 34(10): 2666-2674.
[29] 蔡玉兰, 崔毅. Merton违约距离模型对企业财务困境的预测能力研究——基于离散时间风险模型的实证分析[J]. 预测, 2015, 34(6): 33-38.
[30] 鲍新中, 杨宜. 基于聚类-粗糙集-神经网络的企业财务危机预警[J]. 系统管理学报, 2013, 22(3): 358-365.
[31] Mckeague I W, Sasieni P D. A partly parametric additive risk model[J]. Biometrika, 1994, 81(3): 501-514.
[32] Martinussen T, Scheike T H. Dynamic regression models for survival data[M]. New York: Springer, 2006.
[33] 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.