题名

違約機率預測與極端值

并列篇名

Prediction of Probability of Default and Outlier-Robust Logistic Regression

DOI

10.6545/JFS.2005.13(3).1

作者

沈中華(Chung-Hua Shen);林公韻(Kung-Yun Lin)

关键词

違約機率 ; 羅吉斯模型 ; 穩健迴歸 ; 離群值 ; Probability of Default ; Logit Model ; Robust Regression ; outlier

期刊名称

財務金融學刊

卷期/出版年月

13卷3期(2005 / 12 / 31)

页次

1 - 32

内容语文

繁體中文

中文摘要

本研究建議一個預測這約械率(Probability of Default,以下簡稱PO)的新方法,即使用Logit Model且考慮極端值,在本文為之為Robust Logistic Regression(穩健羅吉斯迴歸),首先,在模型的解釋能力上,Robust Logit Model的Pseudo-R-square值顯著地高於Logit Model,表示使用Robust Logistic Regression能讓模型的解釋能力提升。其次,本研究在樣本內預測的實證結果顯示,不論使用CAP、ROC、KS、Brier Score及交叉分類表,使用Robust Logistic Regression對於整個模型的預測效力會有大幅提升的效果,這表示離群值的存在的確降低核型的預測效力,所以在進行PO的預測時,是必須要考慮昌在群值的影響的,而本研究所提出的這個PO估計方法,的格可以消彌離群值所造成的不良影響,進而提升模型的預測效力。第三,在樣本外的CAP及KS檢定,Robust Logit Model的表現只略優於Logit,而就ROC及Brier Score言,並未得到較佳的預測結果,而使用交叉分類表,則在型一錯誤較佳,在型二錯誤較弱,可能是因為使用Robust Logistic Regression的模型雖然能得到較高的TP%,但卻得到較低的TN%,代表Robust Logistic Regression的模型可以降低型一錯誤,但卻提高了型二錯誤。

英文摘要

A new method for predicting the probability of default (PD) is suggested in this paper. This method, which is named "Robust Logistic Regression" in this paper, is based on Logit Regression but takes outlier into consideration. Firstly, comparing the explainatory power of Robust Logistic Model and Logit Model, the Pseudo-R-square of the former is outstandingly higher than the latter. Secondly, as the empirical study for the insample data shows, CAP, ROC, KS test, Brier Score and classification table all indicate that Robust Logistic Regression can strikingly increase the model's predictive power. It means the existences of the outlier indeedly weaken model's prediction power and the method suggested in this paper can eliminate the negative influences caused by the outlier. Thirdly, as the empirical study for the outsample data shows, Robust Logit Model is better than Logit Model in CAP and KS test but in ROC and Brier Score is not. For the classification table, Robust Logit Model performs better in Type one error but worse in Type two error. It shows Robust Logistic Regression may get higer percentage of TP but lower percentage of TN and means that Robust Logistic Regression can decrease Type one error but increase Type two error.

主题分类 社會科學 > 經濟學
社會科學 > 財金及會計學
参考文献
  1. Altman, E. I.(1968).Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy.Journal of Finance,23,589-609.
  2. Andong Thu,Michael Ash and Robert Pollin(2002).Stock Market Liquidity and Economic Growth: A Critical Appraisal of The Levine/Zervos Model, working paper seriesStock Market Liquidity and Economic Growth: A Critical Appraisal of The Levine/Zervos Model, working paper series,Political Economy Research Institute.
  3. Andreas Cbristmann(1994).Least median of weighted squares in logistic regressionwith large strata.Biomelrika,81,413-417.
  4. Anthony C.,Atkinson and Marco Riani(2001).Regression diagnostics for binomial data from the forward search.
  5. Atkinson, A. C.(1994).Fast Very Robust Methods for the Detection of Multiple Outliers.Journal of the American Statistical Association,89,1329-1339.
  6. Atkinson, A. C.(1985).Plots, transformations and regression.
  7. Bahnson, P. R.,J. W. Bartley(1992).The Sensitivity of Failure Prediction Models to Alternative Definitions of Failure.Advances in Accounting.
  8. Barrett, B. E.,J. B. Gray(1997).Leverage, Residual, and Interaction Diagnostics for Subsets of Cases in Least Squares Regression.Computational Statistics and Data Analysis,26,39-52.
  9. Beaver,W. H.(1966).Financial Ratios as Predictors of Failure.Journal of Accounting Supplement.
  10. Black, F.,M. Scholes(1973).The Pricing of Options and Corporate Liabilities.Journal of Political Economy.
  11. Blum, M.(1974).Failing company discriminate analysis.Journal of Accounting Research,12,72-102.
  12. Booth, P. J.(1983).Decomposition Measures and the Prediction of Financial Failure.Journal of Business Finance and Accounting,10,67-85.
  13. C. H. Shen,C. C. Lee(2005).Some Financial Development Yet Different Economic Growth-but Why?.Joumal of Money.
  14. Cook, R. D.,Weisberg, S.(1982).Residuals and Influence in Regression.
  15. Deakm, E. B.(1972).A Discriminant Analysis of Predictors of Business Failure.Journal of Accounting Research,10,167-179.
  16. Donoho, D. L.,Huber, P. J.(1983).The notion of breakdown point.A Festschrft for Erich L. Lehmann.
  17. Esteban Flores,Jose Garrido(2001).Robust Logistic Regression for Insurance Risk Classification, working paper 01-64.Business Ecnomics Series l3.
  18. Hadi, A. S.,Simonoff, J. S.(1993).Procedures for the Identification of Multiple Outliers in Linear Models.Journal of the American Statistical Association,88,1264-1272.
  19. Haslett(1999).Simple derivation-subset deletion diagnostics.
  20. Jorge R. Sobehart,Sean C. Keenan,Roger M. Stein(2000).Validation Methodologies for Default Risk Models, Moody`s Investors Service.
  21. Jorge R. Sobehart,Sean C. Keenan,Roger M. Stein(2000).Benchmarking Quantitative Default Risk Models: A Validation Methodology.Global Credit Research, Moody`s Investors Service.
  22. Kaplan, R. S.,G. Urwitz(1979).Statistical Models of Bond Ratings: A Methodological Inquiry.Journal of Business,52(2),231-262.
  23. Levine, Ross,Zervos, Sara(1998).Stock markets, Banks, and Growth.American Economic Review,88(3),537-558.
  24. Martin. D.(1977).Earning Warning of Bank Failure: A Logit Regression Approach.Journal of Banking and Finance,1,249-276.
  25. Merton, R. C.(1974).On the Pricing of Corporate Debt: The Risk Structure of Interest Rates.Journal of Finance.
  26. Ohlson, J. T.(1980).Financial Ratios and the Probabilistic Prediction of Bankruptcy.Journal of Accounting Research,18,109-131.
  27. Peter J. Rousseeuw,Annick M. Leroy(1987).Robust Regression and Outlier Detection.
  28. Roger M. Stein(2002).Technical Report #30124, Moody`s KMVTechnical Report #30124, Moody`s KMV,未出版
  29. Rousseeuw P. J.(1984).Least median of squares regression.Journal of the American Statistical Association,79,871-880.
  30. Rousseeuw P. J.(1983).Regression techniques with high breakdown point.The Institute of Mathematical Statistics Bulletin,12,155.
  31. Rousseeuw P. J.,Yohai V. J.(1984).Robust regression by means of S-estimators.Robust and Nonlinear Time Series Analysis.
  32. Rousseeuw P. J,Adreas Christmann, Robustness against separation,outliers in logistic regression(2002).Computational Statistics & Data Analysis,43,315-332.
  33. Scott, W, R.(1997).Financial Accounting Theory.
  34. Taffler, R. J.(1983).The Assessment of Company Solvency and Performance Using a Statistical Model.Accounting and Business Research.
  35. 王懷德(2002)。東吳大學會計研究所。
  36. 江欣怡(2003)。東吳大學國際貿易學系研究所。
  37. 何太山(1977)。政治大學企業管理研究所。
  38. 吳念芳(2003)。國立高雄第一科技大學財務管理研究所。
  39. 吳秉勳(2000)。Detection of Outliers with Data Transformation。
  40. 呂倩如(2002)。On the CAPM from the Views of Robustness and Longitudinal Analysis。
  41. 沈中華。違約機率與博達營收增加。93年8月24日經濟日報。
  42. 沈中華、張家華(2005)。產業違約率及景氣循環。金融風險管理季刊,1(4)
  43. 沈中華、賴柏志、張家華(2005)。總體經濟因素在Basel Ⅱ資本適足率公式的內涵及意義。金融風險管理季刊,1(2)
  44. 卓怡如(1995)。台灣大學財務金融研究所。
  45. 周培如(2004)。國立政治大學經濟學系研究所。
  46. 林妙宜(2001)。國立政治大學金融學系研究所。
  47. 林宓穎(2001)。國立政治大學財政學系研究所。
  48. 林鴻傑(1996)。大葉大學事業經營研究所。
  49. 邱順南(2003)。嶺東技術學院財務金融研究所。
  50. 范少華(2003)。Robust Diagnostic for the Logistic Regression Model With Incomplete Data。
  51. 張大成、劉宛鑫、沈大白。信用評等模型之簡介。中國商銀月刊91年11月號
  52. 張宸豪(2002)。元智大學管理研究所。
  53. 陳明賢(1985)。台灣大學商學研究所。
  54. 陳建賓(2003)。淡江大學財務金融學系研究所。
  55. 陳肇榮(1983)。政治大學企業管理研所。
  56. 曾素娟(1999)。國立成功大學企業管理學系研究所。
  57. 黃小玉(1988)。淡江大學管理科學研究所。
  58. 黃文隆(1993)。東吳大學管理科學研究所。
  59. 黃逸勤(2001)。國立政治大學統計研究所。
  60. 歐再添(2002)。國立台灣科技大學企業管理系研究所。
  61. 潘玉葉(1990)。淡江大學管理科學研究所。
  62. 饒多年(2002)。交通大學經營管理研究所。
被引用次数
  1. Lee, Yung-Hsin,Lee, Tian-Shyug,Chen, Yi-Wen,Chen, I-Fei(2013).A model for enterprise financial distress prediction with cross-section and vertical-section methods.Journal Of Data Analysis,8(2),71-92.
  2. 高惠松(2012)。融合公司治理的信用評等模型:Cubist迴歸樹模型之應用。當代會計,13(2),117-159。
  3. 李沃牆、朱竣平(2008)。信用評等、公司違約率與財務危機預測之探討。真理財經學報,18,33-70。
  4. 連志剛(2013)。無形資產、研發投入與公司財務危機預警模型之研究。人文暨社會科學期刊,9(1),21-33。
  5. 沈中華、毛治文(2009)。退休基金投資策略與股票市場發展。財務金融學刊,17(3),37-72。
  6. 魏裕珍、盧陽正、廖婉茹、張倉耀(2012)。公開新聞之資訊內涵能否增進臺灣企業信用評級慣用指標的預測能力?。台灣金融財務季刊,13(4),27-53。
  7. 張大成、林郁翎(2009)。具多重債務結構企業信用風險管理模式之建構。管理評論,28(4),43-63。
  8. (2010)。公司危機預測:計量模型與變數選取。期貨與選擇權學刊,3(1),57-82。
  9. (2012)。公開新聞之資訊內涵能否增進臺灣企業信用評級慣用指標的預測能力?。台灣金融財務季刊,13(4),27-53。