Una comparacion de la seleccion de los ratios contables en los modelos contable-financieros de prediccion de la insolvencia empresarial

During the last 30 years the growing appearance of quantitative models about insolvency prediction in the financial and accounting literature has awakened a great interest among the specialists and researchers of this field. What in the beginning were a few models with a sole objective, has evolved into a source of constant research. In this paper an insolvency prediction model is formulated through a combination of different quantitative variables extracted from the Annual Accounts of sample firms for the period 1994-1997. Using a stepwise procedure, those variables, which proved to be the most relevant in providing information were selected and analysed. Once we have formulated these models, we looked for an alternative to the previous variables through the use of factorial analysis of main components and it is made a variable selection through this technique. The univariate analysis is applied to both groups of ratios. Lastly, we compared the models obtained and we concluded that although the ratios of previous literature offer better results, the models with the variables of factorial analysis should not be rejected because the causes of insolvency are clearer than in those models that used variables from popularity in literature.