Predictive potential of the bankruptcy global models in the tourism industry

• Autor(a) principal: Del Castillo García, Agustín
• Data de Publicação: 2021
• Outros Autores: Fernández Miguélez, Sergio Manuel
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://tmstudies.net/index.php/ectms/article/view/1512
• topic: Bankruptcy; prediction; tourist firms; artificial neural networks; multi-layer perceptron

The globalization process and the recent economic crises have increased the development of models to identify the factors related to business bankruptcy. As a consequence, previous research has provided bankruptcy prediction models both of a global nature for certain industries or regions of the world, and models focused on certain economic activities. The tourism industry is not immune to this concern and in the previous literature there are certain bankruptcy prediction models, generally focused on hotels or restaurants. However, there are no experiences of global models for tourism companies, although in the bankruptcy prediction literature there is numerous evidence of the advantages offered by global models compared to those focused on single business activity. This study develops a global bankruptcy prediction model capable of predicting with high precision any of the activities carried out in the tourism industry. To this end, a sample of 406 Spanish companies that have developed their activity in three sectors of the tourism industry (hotels, restaurants, and travel agencies) in the period 2017-2019 has been used. This sample includes bankrupt and non-bankrupt corporations and has allowed the comparison between a global model and various focused models applying artificial neural network techniques. The results have confirmed the superiority of the global model and provide different sample selection and cost minimization solutions for bankruptcy prediction modeling in the tourism industry.