Title: A study of corporate failure using data envelopment analysis and quadratic discriminant analysis: the case of the Greek construction sector during the period of the economic crisis

Authors: Apostolos G. Christopoulos; Nikolaos Demiroglou; Ioannis G. Dokas

Addresses: Faculty of Economics, National and Kapodistrian University of Athens, 1 Sofokleous str., (Office 404), 10559, Athens, Greece ' Department of Public Administration, Panteion University of Social and Political Sciences, 12 Kolokotroni, 15127, Melissia, Athens, Greece ' Department of Economics, Democritus University of Thrace, University Campus Komotini, Greece

Abstract: The aim of the present study is to provide a reliable methodology relative to the use of MDA in the construction process of corporate failure prediction models. In the present study, a two-stage analysis is followed. In the first stage, we construct our sample for each sub-period: the pre-crisis period 2006-2008 and the crisis period 2009-2011. The distinction of the total sample in two groups (efficient, non-efficient) per sub-period is realised using DEA method. In the second stage, we develop the QDA and LDA in order to establish a reliable prediction model per sub-period. In more detail, we propose an output-oriented DEA model using six financial ratios from the fields of liquidity, activity and profitability. According to the efficiency score of DEA, we set two groups of firms for each sub-period. In the second stage emphasis is put on the assumptions of the model used. Firstly, we select financial ratios from the categories of capital structure, profitability and operating ratios. Secondly, the necessary normality tests and transformations are incorporated in the proposed methodology in order to apply the MDA.

Keywords: data envelopment analysis; DEA; financial ratios; linear discriminant analysis; LDA; quadratic discriminant analysis; QDA; corporate failure prediction; construction sector; business cycle.

DOI: 10.1504/IJFERM.2019.101293

International Journal of Financial Engineering and Risk Management, 2019 Vol.3 No.2, pp.200 - 221

Received: 05 Oct 2018
Accepted: 12 Mar 2019

Published online: 30 Jul 2019 *

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