La muestra de empresas en los modelos de predicción del fracaso: influencia en los resultados de clasificación

This paper focuses on the development of both failure prediction models on a paired sample and a random sample of small and medium-sized firms with head offices located in the region of Castilla y León (Spain), in order to prove if the predictive power of the developed models is affected by the method used to derive the sample aim of each study. To estimate both models, we consider a set of financial ratios as independent variables in each one, which is first reduced by the application of a principal components analysis. Next, a logistic regression analysis is applied to identify those variables that best explain and predict failure in the two samples, where differences in the significant variables and the classification results are observed, which confirms the influence of the sampling method on the business failure prediction results.