The Hyperbolic Oriented Efficiency Measure as a Remedy to Infeasibility of Super Efficiency Models

Super efficiency models are useful for computing the Malmquist index and in methods for outlier detection, but conventional super-efficiency models may not have feasible solutions for some data sets. We show that a hyperbolic orientation eliminates the infeasibility problem, and permits zero-valued inputs in certain cases. We also demonstrate the computational feasibility of the hyperbolic orientation. Together, these results establish the use of the hyperbolic orientation as a realistic alternative in practice