應用市場資訊於企業危機預警之研究

The Application of Market Information in Predicting Corporate Bankruptcy

10.7047/LRR.200706.0001

張大成；林郁翎；林修逸

企業危機預警 ； 多元區別分析、Logit模型 ； 倒傳遞類神經網路 ； Merton模型 ； Bankruptcy Prediction ； MDA ； Logit Model ； Neural Network ； Merton Model

運籌研究集刊

6卷1期（2007 / 06 / 01）

1 - 18

繁體中文

本文分別使用多元區別分析模型(Multiple Discriminant Analysis, MDA)、Logit模型與倒傳遞類神經網路模型(Neural Network, NN)，建立企業危機預警模型，以Merton(1974)模型的違約距離爲市場基礎變數，將「市場基礎」資訊和傳統「會計基礎」資訊納入其中，以觀察市場資訊對企業危機預警模型的貢獻。研究發現僅使用會計基礎的財務比率爲投入變數的模型，其訓練樣本以Logit模型表現最佳；但在測試樣本中，則三個模型各有優劣。就使用市場基礎的違約距離和會計基礎的財務比率爲投入變數的模型而言，其預測結果則普遍優於僅使用財務比率之模型；在模型正確率上，Logit模型的預測能力優於其他兩個模型，但在型－誤差方面，則以MDA爲佳。故推論市場資訊對於企業危機預警模型之預測能力具有提升效果，且MDA、Logit與NN三個模型中，在模型正確率的指標上以Logit模型的表現最爲優異，而在型－誤差的指標上，則以MDA模型較好故本文除可提供國內銀行業信用風險內部評等模型建立之參考外，研究結果亦可作爲未來相關領域進一步研究之參考依據。

In this paper, Multiple Discriminant Analysis (MDA), Logit Model, and the Neural Network (NN) are used in predicting corporate bankruptcy. We use Merton (1974) model's distance to default as market-based variable, including market-based and accounting-based information, to investigate the contribution of market-based information to the prediction of corporate bankruptcy. The result shows the Logit model has high performance among the models established by using accounting-based data only with the same training samples. Three models have their respective merits. But generally, the model using both market-based data and accounting-based data as input is better than the model using accounting-based data as input in the prediction of corporate bankruptcy. As for the model's accuracy, the Logit model has higher prediction power, but considering type-one error the MDA model is better. In conclusions, the study shows that the models using of market-based information provide an improvement in the prediction accuracy of corporate bankruptcy. Among the MDA, Logit and NN models, we find the Logit model with the highest accuracy in prediction of corporate bankruptcy, and the MDA model has the lowest type I error. This study has also provided useful reference for banks to build credit risk model, and can be explored further for future research.

1. Altman, E.(1968).Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy.Journal of Finance,123(4),589-609.
2. Altman, E.,Katz, S.,M. Schiff (eds.),G. Storter (eds.)(1976).Statistical Bond Rating Classification Using Financial and Accounting Data.Proceeding of the Conference on Topical Resarch in Accounting,New York:
3. Atiya, A.(2001).Bankruptcy Prediction for Credit Risk Using Neural Networks: A Survey and New Results.IEEE Transactions on Neural Networks,12(4),929-935.
4. Working Paper
5. Beaver, W.(1966).Financial Ratios as Predictors of Failure.Journal of Accounting Research,4,71-111.
6. Belkaoi, A.(1980).Industrial Bond Ratings: A New Look.Financial Management,9(3),44-51.
7. Chaveesuk, R.,Srivaree-Ratana, C.,Smith, A.(1997).Alternative Neural Network Approach to Corporate Bond Rating.Journal of Engineering Valuation and Cost Analysis,2(2),117-131.
8. Coats, P.,Fant, L.(1993).Recognizing Financial Distress Patterns Using a Neural Network Tool.Financial Management,22(3),142-155.
9. Ederington, L.(1985).Classification Models and Bond Ratings.Financial Review,20(4),237-262.
10. Working Paper
11. Horrigan, J.(1965).Some Empirical Bases of Financial Ratio Analysis.The Accounting Review,40(3),558-568.
12. Kaminsky, G.(1999).Currency and Banking Crises: The Early Warnings of Distress.IMF Working Paper, No. 178.
13. Kiviluptp, K.(1998).Predicting Bankruptcies with the Self-organizing Map.Neurocomputing,21(1-3),191-201.
14. Lee, K.,Han, I.,Kwon, Y.(1996).Hybrid Neural Network Models for Bankruptcy Predictions.Decision Support Systems,18(1),63-72.
15. Leshno, M.,Spector, Y.(1996).Neural Network Prediction Analysis: The Bankruptcy Case.Neruocomputing,10(2),125-147.
16. McKee, T.,Greenstein, M.(2000).Predicting Bankruptcy Using Recursive Partitioning and a Realistically Proportioned Data Set.Journal of Forecasting,19(3),219-230.
17. Merton, R.(1974).On the Pricing of Corporate Debt: The Risk Structure of Interest Rates.The Journal of Finance,29(2),449-470.
18. Ohlson, J.(1980).Financial Ratios and the Probabilistic Prediction of Bankruptcy.Journal of Accounting Research,18(1),109-131.
19. Ronn, E.,Verma, A.(1986).Pricing Risk-Adjusted Deposit Insurance: An Option-Based Model.The Journal of Finance,41(4),871-895.
20. Shumway, T.(2001).Forecasting Bankruptcy More Accurately: A Simple Hazard Model.Journal of Business,74(1),101-124.
21. 沈大白、張大成、劉宛鑫(2002)。利用類神經網路建構財務危機預警模型。貨幣觀測與信用評等，38，95-102。
22. 張大成(2003)。企業危機預測模型在台灣的應用與比較。台灣銀行季刊，54(4)，147-163。

1. 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。
2. (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。
3. (2013)。財務危機、違約指標、違約距離與系統風險。朝陽商管評論，12(2)，1-32。