Dear Graziella On 22 March 2011 16:08, graziella <graziella.bona...@unical.it> wrote: > I need to introduce new distribution for the error components of a frontier > model. For this reason, I use the package Frontier 4.1,
The software "Frontier 4.1" is not related to R. However, the R package "frontier" is based on "Frontier 4.1". but it includes some bug fixes and improvements and it is much user-friendlier. Therefore, I recommend to use the R package "frontier" rather than "Frontier 4.1". > but I need to > introduce other distribution function than Normal and Half-Normal (that are > those given by default). Unfortunately, the "R" package "frontier" (as well as "Frontier 4.1") can use only the half-normal and the truncated normal distributions for modelling inefficiency. Implementing further distributions for the inefficiency term is on my to-do list but it has only a mediocre priority. Hence, if you want to use other distributions, you have to implement them yourself either in the (FORTRAN) source code of the "frontier" package or your could start from scratch, write the likelihood functions (and gradient functions) and use, e.g., the maxLik() package for maximising the likelihood. The source code of the "frontier" package is hosted on R-Forge. I am sorry that I and the "frontier" package cannot be more helpful right now. Best wishes from Copenhagen, Arne -- Arne Henningsen http://www.arne-henningsen.name ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
