The weights of the

Z-Score model are almost similar to the regression analysis.

It was concluded that both

Z-score model and O-score model can be used to predict the bankruptcy risk.

3.3 Development of the

Z-Score Model. Given the estimation sample of 36 performing loans and 4 non-performing loans, the multivariate linear discriminant functions have been estimated to obtain Z scores that will help to predict the credit risk involved for new loan proposals.

Other suggestions for future study include comparing neuropsychology raw score models with

z-score models that adjust for more than age (e.g., education, gender, and race), examining the role of cognition in predicting other outcome variables (e.g., discharge disposition, bowel and bladder continence, and length of rehabilitation stay), and examining other possible predictors of functional ability (e.g., age, presence of depression, and type of cognitive deficit).

The original

Z-score model was revised and modified several times in order to find the scoring model more specific to a particular class of firm (for a detailed presentation of the different developments please refer to Altman and Narayan, 1997 and Altman, 2000).

We employ a widely used accounting ratio-based

z-score model as a proxy for default risk as with Dichev (1998), Griffin and Lemmon (2002), and Ferguson and Shockley (2003).

Such a model is also known as Zeta function or a

Z-score model.

Techniques used for analyzing the data was

Z-score model, mean, standard deviation, co-efficient of variation and graphs.

As such, the

Z-Score model, which was originally intended to track the possibility of domestic manufacturing company bankruptcies, was updated and re-released in the mid-1990s.

By means of his analytical method he got the following formula known as the Altman's bankruptcy predictive model or the

Z-Score model, which is used for companies listed at the capital market: