Nonparametric, Stochastic Frontier Models with Multiple Inputs and Outputs

Stochastic frontier models along the lines of Aigner et al. are widely used to benchmark firms’ performances in terms efficiency. The typically fully parametric, with functional form specifications for as well both noise and inefficiency processes. Studies such Kumbhakar have attempted relax some restrictions parametric models, but so far all approaches limited a univariate response variable. Some (e.g., Simar Zelenyuk; Kuosmanen Johnson) proposed nonparametric estimation directional distance functions handle multiple inputs outputs, raising issues endogeneity that either ignored or addressed by imposing restrictive implausible assumptions. This article extends methods developed Hafner allow outputs an almost framework while avoiding problems. We discuss properties resulting estimators, examine their finite-sample performance through Monte Carlo experiments. Practical implementation method is illustrated using data on U.S. commercial banks.