Altman Z-score
The 1968 published Z- factor model of Altman ( Altman 's Z -Score ) is the first multivariate insolvency prediction for business at all. For individuals, however, multivariate analysis methods have already been applied at this time. In the finance literature, it is still often used as a benchmark to assess the predictive power of rating systems and even forms the basis of some commercial rating process., There are currently three versions of the Z -factor method, the Z, Z ' and Z "model are referred to. Newer, but empirically well-founded studies consistently show that none of the model versions of the Z -factor method is well suited for the prediction of corporate bankruptcies. More appropriate indicators methods are publicly available.
The different model versions of the Z- factor method
The Z- factor method
The Altmansche Z -factor method was parameterized by means of multivariate linear discriminant analysis and assigns the company to be assessed in two groups ( " expected solvent" vs. " Likely insolvent" ). In practice, the z-factor value is also interpreted ordinal. Based on a very small sample of 33 bankrupt and 33 solvent companies succeeded in Altman (1968 ) for a forecast horizon of one year, 31 of the bankrupt and 32 non- bankrupt companies to classify correctly. The assignment was based on the discriminant function:
Legend:
- = (Current assets - current liabilities ) / total assets,
- = Retained earnings / total assets,
- = Earnings before interest and taxes (EBIT ) / total assets,
- = Market value of equity / total liabilities,
- = Sales / total assets
Companies with a Z -factor value of less than 1.81 considered according to this method on single year as highly at risk of insolvency, companies with a Z-score value greater than 2.67 as safely.
The Z'- factor method
To apply the Z- factor method to non - listed companies, was replaced in a later model revision in the definition of key figures for the market value of equity by the book value and the re-estimated all coefficients:
Legend:
- = Book value of equity / total liabilities
For the other variables, see above
The Z 'factor method
Since the measure was considered to be industry- dependent, it has been removed as part of a third version of the model, whose aim was to apply the model to companies outside the manufacturing sector. The coefficients of the remaining variables were re-estimated:
Legend: See Z'- factor method
In this model, a company must be a Z ' value of less than 1.10 as at risk of insolvency.
Results of recent empirical studies
In a Metauntersuchung of ten empirical studies in which several thousand companies were included, the forecast performance of the various Z- factor method were measured. Although the Z -factor model was originally adapted to U.S. public companies with data of ( from today's perspective ) 40 to 60 years old financial statements, the univariate forecast performance of the indicators used are ( with the exception of the measure X5, which from the Z " was - factor method is no longer used ) even with current financial statements of international or German or Austrian SMEs not bad .. bad however is the calibration of the coefficients. except for the original study Altman (1968 ) and his later studies, the empirical results consistently very poor. forecasting performance of the three Z- factor method This is true not only in comparison to the developed by the respective authors themselves models - but even compared to the ( univariate ) forecast performance of individual indicators, such as the return on assets (net income / total assets) or the equity ratio (!) (equity / total assets).
Swell
Left
- Altman Z -Score Free application to Android OS.
- Free online Altman Z-score calculator