Testing scale efficiency in DEA models: a bootstrapping approach

This paper presents a method for estimation and test of firm-specific scale efficiency using data envelopment analysis (DEA). The use of a simple bootstrap algorithm is proposed to perform firm-specific inference for scale efficiency. An empirical application using Swedish hospital data is provided. It is found that about 40% of the departments were scale efficient according to the original results. However, for about one-third of those departments the hypothesis of scale efficiency could be rejected.