Prediction of tourism zombie companies using artificial intelligence algorithms and accounting data

Abstract

This paper addresses the problem of the prediction of zombie companies. Despite their relevance, financial issues represent an area scarcely considered in relation to the tourism industry: the topic has so far remained totally unexplored. Zombie firms cause great damage to the sector in which they compete. However, there is no consensus regarding what actions should be applied to them due to the negative consequences of maintaining or liquidating them. In view of this, the key issue is to prevent them from entering the zombie state. It would therefore be useful to have a methodology for predicting years in advance when a company will become a zombie firm. For this purpose, this article has used different artificial intelligence algorithms applied to easily obtained accountant data. Utilizing a dataset of 356 Spanish small and medium-sized enterprises in the tourism sector, 78.4% of predictions have been correct.