CORPORATE BANKRUPTCY PREDICTION USING ALTMAN’S Z-SCORE MODEL: THE EFFECT OF TIME AND METHODOLOGY ON ACCURACY OF THE MODEL
CORPORATE BANKRUPTCY PREDICTION USING ALTMAN’S Z-SCORE MODEL: THE EFFECT OF TIME AND METHODOLOGY ON ACCURACY OF THE MODEL

Summary/Abstract: The stability of the bankruptcy prediction models and their predictive power is an empirical question in the area of bankruptcy prediction. Altman’s Z-score model is pioneer study and widely used to predict the bankruptcy. The aim of the paper is to examine sensitivity of the Altman’s model to variations in time and methodology. For this purpose a sample of 74 Indian manufacturing companies equally divided into defaulted companies and non-defaulted companies from 2011-2015 is used. The study investigates the time varying effect on the accuracy of the model by re-estimating the model coefficients using the recent data. In place of Multiple Discriminant Analysis (MDA) logistic regression is employed to examine the effect of change in methodology on the accuracy of the model. The findings of the study reveal that the overall correct classification rate for Altman’s model in 2015 found to be only 66.21% which even decreases further in the years prior to 2015. This overall correct classification rate increases to 81.10% when the model is re-estimated by using the recent data and further to 87.83% when logistic regression is applied to re-estimate the model. The research findings strongly confirm that the Altman’s model is sensitive to time period variation and change of methodology. The findings of the study suggest that the model coefficients should be re-estimated based on recent data using logistic regression to have better accuracy in bankruptcy prediction.

• Details
• Contents