Title: Failure prediction models: development and comparison between the multivariate discriminant analysis and the support vector machine for Tunisian companies

Authors: Fayçal Mraihi; Inane Kanzari

Addresses: Higher School of Economics and Business Sciences of Tunis, 4, Rue AbouZakaria El Hafsi – 1089 Montfleury – Tunis, Tunisia ' Higher School of Economics and Business Sciences of Tunis, 4, Rue AbouZakaria El Hafsi – 1089 Montfleury – Tunis, Tunisia

Abstract: In this study, we try to develop a model that would predict corporate default using a multivariate discriminant analysis (MDA) and a support vector machine (SVM). The two models are applied on the Tunisian context. Our sample consists of 212 companies operating in different industries, of which 106 are 'performing' companies and 106 are 'failing' companies, observed over the 2005-2010 period. The results of the use of a battery of 87 ratios showed that 16 ratios can build the model and that liquidity and solvency have more weight than profitability and management in predicting distress. Despite the slight superiority of the results provided by the SVM model, on the control sample, the results provided by the two models are good either in terms of correct classification percentage or in terms of the stability of discriminating power over time and space.

Keywords: distressed firms; forecasting model; multivariate discriminant analysis; MDA; support vector machine; SVM.

DOI: 10.1504/IJESB.2021.115882

International Journal of Entrepreneurship and Small Business, 2021 Vol.43 No.3, pp.411 - 437

Received: 19 Feb 2018
Accepted: 28 Jan 2019

Published online: 05 Jul 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article