Posts tagged ‘Zeta Model’

Altman Z-Score / Zeta Model of Credit Analysis

The Altman zeta model is a formula that predicts whether a company will go bankrupt within the following two years. Generally speaking, any company with a score of 1.8 or lower is in trouble, while a score of 3.0 or higher means the company is safe. If the score is between 1.8 and 3.0, then it is in a danger area; depending on the company’s policies and market fluctuations it could go either way. Given the value of the five variables for a given firm, a Z-score is computed.

Z = 1.2X1 + 1.4X2+ 3.3X3 + 0.6X4 + 1X5

Z: Z score
X1: Working capital/Total assets (in decimal)
X2: Retained earnings/Total assets (in decimal)
X3: Earnings before interest and taxes/Total assets (in decimal)
X4: Market value of equity/Total liabilities (in decimal)
X5: Sales/Total assets (number of times)

Extensive testing has found this formula to be very accurate, as high as 90% in some cases. It works by weighting various aspects of a company’s assets and earnings in order to generate a final score. The original model was designed for certain manufacturing companies, but it has since expanded to cover other areas. In Altman’s original testing, the zeta model was found to have 72% accuracy with a false positive chance of 6%. These numbers made it far more accurate than any other model currently used. Since the initial testing, the model has been put up against a huge number of cases, and it is believed to have an 85% to 90% accuracy rate with a false positive rate of 15 to 20%, still far above other models.

Below is the calculated Z-score based on Zeta model dictated above of Few Indian Real Estate Companies:

Note: Oberoi has zero debt but for some rational Z-Score I have taken a Debt figure of Rs.1crore.

There are said to be few drawback as well:

  • There are only two extremes or say two states i.e. Default or no Default
  • The second drawback is the constant factor weights which few analyst may agree with and few disagree.
  • The model only considers five fundamental quantifiable variables. No Weight for Qualitative information.

Above note is a result of publicly available information!!

More about Kush Sonigara on Google+

Tag Cloud