Método Análisis Envolvente de Datos y Redes Neuronales en la Evaluación y Predicción de la Eficiencia Técnica de Pequeñas Empresas Exportadoras

Abstract

In this research, a method was developed to evaluate and predict the efficiency of small exporting companies taking as input or asset variables the total assets, equity, total liabilities, operating expenses, sales costs and as output or result variables. net sales, net income and operating income. For this, the envelopment data analysis was used in the evaluation of the efficiency, the discriminant analysis in the evaluation of the classification of efficient and inefficient companies and the artificial neural networks to evaluate its capacity of classification prediction in 90 companies of the sector of the city of Barranquilla-Colombia. The results allowed to classify the companies according to level of efficiency showing an average technical efficiency of 41.38% of the sector with 11 representative companies of efficiency. The results show the relevance of the proposed methodology to correctly classify and forecast technical efficiency in small exporting companies.