112
Views
8
CrossRef citations to date
0
Altmetric
Original Articles

Análisis de sensibilidad temporal en los modelos de predicción de insolvencia: una aplicación a las PYMES industrials

Analysis of temporal sensitivity in the insolvency prediction models: an application to the industrial small business

, &
Pages 85-111 | Received 01 Dec 2005, Accepted 20 Sep 2007, Published online: 15 Jan 2014
 

RESUMEN

La capacidad predictiva de los modelos que miden el riesgo de insolvencia puede mejorarse si se consideran aspectos como el desarrollo de modelos sectoriales, la incorporación de variables no financieras y el horizonte temporal. En este trabajo se ha utilizado una muestra de 364 empresas con información relativa al trienio 1998–2000, y se han obtenido tres funciones logísticas para medir el riesgo de que una PYME industrial incurra en un proceso concursal con tres, dos y un año de antelación. En general, los modelos obtenidos son muy similares en cuanto a variables significativas, alcanzando un mayor grado de acierto global el que predice dos años antes. Como a priori no se conoce en qué momento se producirá la insolvencia, se han cruzado modelos, comprobándose que no existen diferencias significativas en el error que se cometería aplicando los modelos obtenidos en fechas distintas de las consideradas para su estimación.

ABSTRACT

The predictive capacity of the models that measure the insolvency risk can be improved if aspects as economic activity, non-financial variables and temporary horizon are considered. In this work, a sample of 364 companies with information relative to 1998–2000 has been used, and three logistic functions have been obtained to measure the risk of which an industrial small and medium business incurs in legal bankruptcy with three, two and one year in advance. In general, the models obtained are very similar with respect to significant variables, reaching a greater degree of global success the one that predicts two years before. Since we do not know when the insolvency will take place, we have crossed the models, verifying that it does not exist significant differences in the error that it would make applying the obtained models in different samples to those considered for its estimation.

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.