Ana Isabel Irimia Diéguez

Utilización de técnicas paramétricas y no paramétricas para la elaboración de un modelo de predicción de quiebra en las microempresas

In this paper the question, becoming a pioneering work in this field, while the methodology used and the sector to which it applies is addressed. Based on financial and non–financial variables that have been used with some success in predicting corporate bankruptcy in general, we try to determine which ones are affecting more to microenterprise, working a hybrid model by combining logistic regression, parametric approach, a nonparametric technique and learning derived from the theory of Rough Sets. The application of the model developed a micro database shows that our hybrid model improves outcomes, to which is attached a faster convergence that each method implemented in isolation. Moreover, the convenience of the introduction of non financial and macroeconomic variables that complement the information provided by financial ratios to predict bankruptcy is found.