A tourist in the economics of tourism: Reflections on nonparametric estimation of stochastic frontier models

The past two decades has seen tourism research embrace frontier methods to assess efficiency in tourism. Generically these methods are split into two different camps, deterministic and stochastic frontier models. Deterministic models are by their nature nonparametric and do not require distributional assumptions. Stochastic models allow for noise to enter the frontier, typically at the expense of a variety of potentially unverifiable assumptions. The primary objective of this short survey is to alert readers to the many advances that have been made to eliminate or mitigate various parametric assumptions in the stochastic frontier model.