Practical consistent estimation of the structural parameters of true fixed-effects stochastic frontier model

True fixed-effects stochastic frontier models are employed in panel data settings to separate time-invariant heterogeneity from efficiency effects. These models have some desirable properties, but the estimation of their structural parameters is hindered by the incidental parameter problem, which may be severe for settings with a large number of short panels. Some consistent estimators have been recently proposed in the econometric literature, but they are rather involved and overly complex to implement. Here we propose an alternative estimator, which has optimality properties while being computationally simple. The proposal results from the application of the equivariance property of maximum likelihood estimation in group families, and it provides a consistent estimator regardless of the size of the panel. The solution covers a broad range of stochastic terms, and it does not require any simulation. TheTMB R package for automatic differentiation is employed to obtain a scalable implementation.