| Popis: |
Skewness is an intrinsic characteristic in Stochastic Frontier Analysis (SFA), where it is used as a measure of technical inefficiency. We discuss the use of skew normality in SFA. We consider stochastic frontier analysis in the common setting Normal + Truncated Normal with uncorrelated errors, as well as the case with correlated errors. In this last case we show the connection between the SFA model and the Closed Skew Normal as discussed in Gonzalez-Faŕias, et al (2003). We end with the proposal of a model for stochastic frontier analysis with elliptical errors. |
