Predicting probability of default of Indian corporate bonds: logistic and

Z-score model approaches.

This study is an effort to identify the best financial and accounting variables which preeminently discriminate bankrupt companies form non-bankrupt companies, to develop

Z-Score model through a procedure identical to that of Altman's (1968) using multivariate statistical technique of discriminant analysis and logit model to provide a reliable tool as Internal rating bases (IRB) model in Pakistani context to banking industry for the credit evaluation of corporate loan applicants.

Rayalaseema and Muhammad (2012) reviewed literature and compared different bankruptcy prediction models and revealed that

Z-score model is a very powerful tool to predict bankruptcy.

Bandyopadhyay (2006) developed

Z-score model for Indian corporates by using working capital/total assets, cash profits/total assets, total non-current and current borrowings/total assets, operating profit/total assets, and sales/total assets ratios, and compared the results with Altman's 4 and 5 ratios models, and found his new Z-score had higher prediction accuracy both for solvent and defaulted firms.

Regression analyses were performed for both RBANS raw score and age-adjusted

z-score models. Patient age was intentionally not used as a covariate in order to compare models that used age-adjusted z-scores against models that did not adjust for age.

Can Altman

Z-Score Models Predict Business Failures in Greece.Research Journal of International Studies, 12.

Altman-Nammacher (1987) suggest that a modified

Z-score model used to predict bankruptcy can also be used to predict default for these high-yield bonds or as a gauge of changes in credit quality.

Note that if the book value variation of the

Z-score model was used instead (substituting the book value of equity of $156.1 for the market value of $242.8), the Z-score would instead be calculated as Z = 0.717 (0.2591) + 0.847 (0.4372) + 3.107 (0.1638) + 0.420 (0.6384) + 0.998 (1.2142) = 2.41, which is considerably below the 3.0 threshold.

The early approach of credit scoring methodology is commonly structured along the lines of Altaian's (1968)

Z-score model. Altaian's

Z-score model is an application of multivariate discriminant analysis (MDA) in credit risk modelling.

Altman's

z-score model and those like it typically control for balance sheet liquidity.

Altman developed his

Z-score model by using manufacturing companies that filed a bankruptcy petition from 1946 to 1965 (Edward I.