Regresar

Predicting growth of micro and small business using neural networks

Abstract:

The study of the future behavior of micro and small enterprises is an important issue for owner-managers related to businesses, the region where they are inserted and the institutions responsible for studying and supporting these activities. In spite of the aforementioned, these kinds of studies are not always carried out at the Ecuadorian context, or the techniques and competences are not available to make them. This work describes the case of an application of neural networks to predict the future economic behavior of the studied businesses. The application is tested on real-world data, related to 7,169 micro and small Ecuadorian businesses. The trained network allows classify correctly the 85.6% of the cases analyzed predicting its economic growth or decrease. This result shows that the neural networks model obtained can be very successful in learning and estimating the growth or decrease tendency of a micro and small businesses.