基于Logistic与Fisher的上市公司财务困境判别模型比较研究
|
摘要
以我国股权分置改革后A股制造业公司为研究对象,采用大样本法选取样本,选用财务困境发生前2年内8个季度的财务数据进行分析,对财务困境公司与正常公司的财务指标分别进行Fisher判别分析、线性logistic回归和非线性logistic回归分析。研究结果表明,我国上市公司财务困境发生前2年的季度数据对公司财务困境是否发生具有预测效力;总资产净利润率、营运资金比率和成本费用利润率三个指标对财务困境的预测能力较强;非线性logistic模型的误判率最低,线性logistic模型其次,Fisher模型的误判率最高。
This paper chooses China's A-share manufacturing companies after the reform of non-tradable shares as the research object,with the use of large sample method to select the samples. The 8 quarterly data in the first two years before financial distress are selected to make a regression analysis of the ST companies and normal companies respectively based on Fisher model,linear Logistic model and non-linear Logistic model. The results show that the quarterly data in the first two years before financial distress has an effect of prediction to the occurrence of financial distress; the net profit margin,working capital ratio and cost margins have better prediction to the financial distress; and the error rate of non-linear Logistic model is the lowest,the rate of the linear Logistic model is in the second place,and the error rate of the Fisher model is the highest.
This paper chooses China's A-share manufacturing companies after the reform of non-tradable shares as the research object,with the use of large sample method to select the samples. The 8 quarterly data in the first two years before financial distress are selected to make a regression analysis of the ST companies and normal companies respectively based on Fisher model,linear Logistic model and non-linear Logistic model. The results show that the quarterly data in the first two years before financial distress has an effect of prediction to the occurrence of financial distress; the net profit margin,working capital ratio and cost margins have better prediction to the financial distress; and the error rate of non-linear Logistic model is the lowest,the rate of the linear Logistic model is in the second place,and the error rate of the Fisher model is the highest.
引文
[1]Kurbat M,Korbalev I.Methodology for testing the level of the EDF credit measure[Z].White Paper,Moody’s KMV,2002.
[2]Crosbie P,Bohn J.Modeling default risk[Z].White Paper,Moody's KMV,2003.
[3]Dwyer D,Korablev I.Power and level validation of Moody’s KMV EDF credit measures in North America,Europe and Asia[Z].White Paper,Moody’s KMV,2007.
[4]索贵彬,赵国杰.现代信用风险度量模型的适用性[J].西北农林科技大学学报:社会科学版,2006(1):68-72.
[5]郭敏.商业银行信用风险度量模型简介及思考[J].上海金融,2007(2):49-51.
[6]朱欣凤.现代信用风险度量模型在我国商业银行的适用性分析[J].现代商业,2008(24):41.
[7]Altman E I.Financial ratios,discriminant analysis and the prediction of corporate bankruptcy[J].The journal of finance,1968,23(4):589-609.
[8]Beaver W H.Financial ratios as predictors of failure[J].Journal of Accounting Research(Supplement),1966(41):71-111.
[9]Ohlson J A.Financial ratios and the probabilistic prediction of bankruptcy[J].Journal of accounting research,1980,18(1):109-131.
[10]Odom M D,Sharda R.A neural network model for bankruptcy prediction[C]∥Neural Networks,1990 IJCNN International Joint Conference on IEEE,1990:163-168.
[11]Bryant S M.A case-based reasoning approach to bankruptcy prediction modeling[J].Intelligent Systems in Accounting,Finance and Management,1997,6(3):195-214.
[12]Laitinen E K,Laitinen T.Bankruptcy prediction:application of the Taylor's expansion in logistic regression[J].International Review of Financial Analysis,2001,9(4):327-349.
[13]吴世农,卢贤义.我国上市公司财务困境的预测模型研究[J].经济研究,2001(6):46-55.
[14]王春峰,万海晖,张维.商业银行信用风险评估及其实证研究[J].管理科学学报,1998(1):68-72.
[15]彭建刚,屠海波,何婧,等.有序多分类logistic模型在违约概率测算中的应用[J].财经理论与实践,2009(4):2-7.
[16]于立勇,詹捷辉.基于Logistic回归分析的违约概率预测研究[J].财经研究,2004(9):15-23.
[17]杨德勇,马若微.现代期权定价理论框架下的财务困境解释与实证检验[J].财贸经济,2009(4):33-37.
[18]刘芳.基于面板Logistic模型的上市公司财务困境实证分析——以装备制造业为例[J].经济视角(下),2011(9):80-82.
[19]徐永智,秦培培.基于Logistic回归的中国创业板上市公司信用风险度量研究[J].经济论坛,2013(2):75-79.
[20]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.
[2]Crosbie P,Bohn J.Modeling default risk[Z].White Paper,Moody's KMV,2003.
[3]Dwyer D,Korablev I.Power and level validation of Moody’s KMV EDF credit measures in North America,Europe and Asia[Z].White Paper,Moody’s KMV,2007.
[4]索贵彬,赵国杰.现代信用风险度量模型的适用性[J].西北农林科技大学学报:社会科学版,2006(1):68-72.
[5]郭敏.商业银行信用风险度量模型简介及思考[J].上海金融,2007(2):49-51.
[6]朱欣凤.现代信用风险度量模型在我国商业银行的适用性分析[J].现代商业,2008(24):41.
[7]Altman E I.Financial ratios,discriminant analysis and the prediction of corporate bankruptcy[J].The journal of finance,1968,23(4):589-609.
[8]Beaver W H.Financial ratios as predictors of failure[J].Journal of Accounting Research(Supplement),1966(41):71-111.
[9]Ohlson J A.Financial ratios and the probabilistic prediction of bankruptcy[J].Journal of accounting research,1980,18(1):109-131.
[10]Odom M D,Sharda R.A neural network model for bankruptcy prediction[C]∥Neural Networks,1990 IJCNN International Joint Conference on IEEE,1990:163-168.
[11]Bryant S M.A case-based reasoning approach to bankruptcy prediction modeling[J].Intelligent Systems in Accounting,Finance and Management,1997,6(3):195-214.
[12]Laitinen E K,Laitinen T.Bankruptcy prediction:application of the Taylor's expansion in logistic regression[J].International Review of Financial Analysis,2001,9(4):327-349.
[13]吴世农,卢贤义.我国上市公司财务困境的预测模型研究[J].经济研究,2001(6):46-55.
[14]王春峰,万海晖,张维.商业银行信用风险评估及其实证研究[J].管理科学学报,1998(1):68-72.
[15]彭建刚,屠海波,何婧,等.有序多分类logistic模型在违约概率测算中的应用[J].财经理论与实践,2009(4):2-7.
[16]于立勇,詹捷辉.基于Logistic回归分析的违约概率预测研究[J].财经研究,2004(9):15-23.
[17]杨德勇,马若微.现代期权定价理论框架下的财务困境解释与实证检验[J].财贸经济,2009(4):33-37.
[18]刘芳.基于面板Logistic模型的上市公司财务困境实证分析——以装备制造业为例[J].经济视角(下),2011(9):80-82.
[19]徐永智,秦培培.基于Logistic回归的中国创业板上市公司信用风险度量研究[J].经济论坛,2013(2):75-79.
[20]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.