fepsfrontieR: Estimating Fixed-Effects Panel Stochastic Frontier Models

In clemenshaerder/fepsfrontieR: Estimation of fixed-effect panel stochastic frontier models by MLE

Description

A package for estimation of fixed-effect panel stochastic frontier models by within or first-difference model transformation as proposed by Wang & Ho (2010). Maximum Likelihood Estimation is performed using nlminb( ). Standard Errors are provided using the Hessian Matrix or by individual Bootstrapping. The fepsfrontieR package provides the main function sfmfep and a supplementary data generation function SFM.generate

sfmfep

The sfmfep functions is the main function of the package fepsfrontieR. sfmfep can be either used to estimate model parameters based on a data input, or to fit a prespecified model. Estimation is performed by applying a within or first-difference transformation of the data. One can optionally choose Bootstrapping to calculate the estimates and the associated standard errors. Otherwise, the Standard Errors are obtained from the hessian matrix. The estimation of the parameters is numerically done using the optimizer nlminb ( ). Starting points for the optimizer can be provided. Otherwise, OLS estimators are used as starting points. Finally, the inefficency index & fixed-effects are calculated. A S3 object is returned.

SFM.generate

The SFM.generate function generates a balanced fixed-effects stochastic frontier model with exponential inefficencys. The amount of specified betas & deltas create associated explenatory varibles & inefficeny determinants.

clemenshaerder/fepsfrontieR documentation built on May 22, 2019, 3:43 p.m.