The objective of this study is to develop probability models for the prediction of financial failure of
Greek industrial firms, based on a set of ratios estimated from published financial statements. A sample of
failed and non-failed firms was selected and financial ratios were computed in order to form a set of
independent variables to be included in the models. Using both Linear Probability and Logit models and
following a factor analysis to select useful ratios, a set of probability models were identified which proved
to be significantly reliable in predicting financial failure, both in the estimation sample and in a separate
validation sample, up to four years before its occurrence. The models were also consistently accurate in
distinguishing financially healthy firms from those facing serious financial difficulties. Differences in the
predictive ability of the two methods employed have not been found significant.
(EN)
University of Peiraeus
