Predicting Corporate Failure of Malaysias Listed Companies: Comparing Multiple Discriminant Analysis, Logistic Regression and the Hazard Model Nur Adiana Hiau Abdullah Universiti Utara Malaysia, Associate Professor Faculty of Finance and Banking, Universiti Utara Malaysia Sintok 06010, Kedah, MALAYSIA E-mail: firstname.lastname@example.org Tel: 00-604-9286464/006013 5306566; Fax: 00-604-9286406 Abd. Halim Universiti Utara Malaysia Hamilton Ahmad Universiti Utara Malaysia Rohani Md. Rus Universiti Utara Malaysia Abstract This study compares three methodologies for identifying financially distressed companies, multiple discriminant analysis (MDA), logistic regression and hazard model. In a sample of 52 distressed and non-distressed companies with a holdout sample of 20 companies, the predictions of the hazard model were accurate in 94.9 % of the cases examined. This was a higher accuracy rate than generated by the other two methodologies. However, when the holdout sample is included in the sample analyzed, MDA had the highest accuracy rate at 85%. Among the ten determinants of corporate performance examined, the ratio of debt to total assets was a significant predictor of corporate distress regardless of the methodology used. In addition, net income growth was another significant predictor in MDA, whereas the return on assets was an important predictor when the logistic regression and hazard model methodologies were used. Keywords: Bankruptcy, Multiple Discriminant Analysis, Logistic Regression, Hazard Model JEL Classification Codes: G33, C51
The sudden currency crisis in 1997 has thrown many financially strong companies out of business. All because they were not able to face the challenges and the unexpected changes in the economy. The growing economy suddenly became an alien to them when depression took place in a split second. As a
International Research Journal of Finance and Economics - Issue 15 (2008)
result, many companies were forced into bankruptcy or became a financially distressed company when they were not able to pay their financial obligations due to inadequate cash flows. Looking at the above situation, it is important to understand the reasons behind the collapse of a company. Knowing these reasons might hinder a company from being financially distress and early actions could be taken as a precaution. Studies in Malaysia have looked into this area, and have used models such as the multiple discriminant analysis (MDA), the logit model or a combination of both models. However, this study takes a different approach where a comparison of three modelsMDA, logistic regression and hazard modelis implemented. The motivation for this study arises from the arguments made by several authors who claimed that MDA suffered from serious drawbacks. Some of these drawbacks were regarding the assumptions of similar variance covariance matrices and linear distributions of independent variables that might lead to invalid results. Logit, on the other hand, uses data averages where a healthy company is given a value of 0 and a distressed company is given a value of 1. Hence, the logit model treats bankrupt companies as if they were bankrupt ever since their inception. A hazard model is able to overcome this problem by examining all firm year observations. These companies would only be assigned as a distressed company in the year they became problematic or distress; otherwise, they are being treated as a healthy company. In comparing the MDA, logistic and hazard models by using a US data set, Shumway (2000) claimed that the hazard model was more reliable and accurate in predicting distress or bankrupty. Based on the above arguments, we are trying to examine the outcome from these different techniques, and to determine which variables appear...