Predicting Corporate Bankruptcy in Pakistan A Comparative Study of Multiple Discriminant Analysis (MDA) and Logistic Regression

ASAD ALI JAFFARI

Abstract


Purpose:- The aim of the study was to predict corporate bankruptcy in an emerging market like Pakistan by employing two statistical methods which are Discriminant Analysis (MDA) and Logistic Regression (Logit). It was also aimed at identifying the predicting accuracies of these statistical methods.

Methodology:- This study had examined 35 bankrupt and 35 non-bankrupt companies which belongs to sector (non-financial) of Pakistan listed at KSE (Karachi Stock Exchange) over the period of seventeen years i.e. 1996 to 2012. Here, we had compared the accuracy and predictive ability of two statistical methods which are Discriminant Analysis (MDA) and Logistic Regression (Logit) and was expecting that Logistic Regression (Logit) accuracy and predictive ability will supercede Multiple Discriminant Analysis (MDA) accuracy and predictive ability.

Findings:- The results have proved that Logistic Regression accuracy and predictive ability (80%) is better than the accuracy and predictive ability (78.6%) of Multiple Discriminant Analysis (MDA). It is proved that both the models identified the same amount of predictors for bankruptcy prediction. The variables identified by Logistic Regression are Shareholder’s Equity / Debt (Book Value), EBIT / Current Liabilities, Retained Earnings / Total Assets and variables identified by Multiple Discriminant Analysis (MDA) are EBIT / Current Liabilities, Sales / Total Assets and Sales / Quick Assets which have shown significant contribution towards bankruptcy prediction.

Originality/Value:-This paper had revealed the accuracy and predictive ability of two statistical methods which are Discriminant Analysis (MDA) and Logistic Regression (Logit) employed in this study and has shown us the better model i.e. Logistic Regression which needs to be used in future for prediction of corporate bankruptcy in Pakistan.

Limitations:- This study had used the small sample size and the focus was at Non-Financial Sector only. So it may be extended to other sectors and to other developing countries of the world.

Practical Implications:- This study will be beneficial for managers and investors.


Full Text: PDF
Download the IISTE publication guideline!

To list your conference here. Please contact the administrator of this platform.

Paper submission email: RJFA@iiste.org

ISSN (Paper)2222-1697 ISSN (Online)2222-2847

Please add our address "contact@iiste.org" into your email contact list.

This journal follows ISO 9001 management standard and licensed under a Creative Commons Attribution 3.0 License.

Copyright © www.iiste.org