Forum: help
Monitor Forum | 
Start New Thread
| RE: Stochastic frontier by true fixed effects (TFE) Greene, 2005 [ Reply ] By: Arne Henningsen on 2022-07-30 12:13 | [forum:49427] |
|
Hi Hernan I give an example in my response in this thread that I gave on 2016-08-24 10:11. Best wishes, Arne |
|
| RE: Stochastic frontier by true fixed effects (TFE) Greene, 2005 [ Reply ] By: Hernan Borrero on 2022-07-27 15:52 | [forum:49425] |
|
Hi dear Arne; according to the description of the Frontier package, it only run the Battese & Coelli (1992) and Battese & Coelli (1995) models. It does not run the Greene (2005) True Fixed Effects (TFE) model. Can you help me investigate how to run the TFE Greene model in R??? Hernan Borrero |
|
| RE: Stochastic frontier by true fixed effects (TFE) Greene, 2005 [ Reply ] By: Arne Henningsen on 2016-08-24 09:33 | [forum:43468] |
|
Hi Luca The "within" transformation only works for simple linear models and cannot be used for estimating SFA models (unless one adjusts the estimation procedure, which then becomes so complex that almost nobody uses this). /Arne |
|
| RE: Stochastic frontier by true fixed effects (TFE) Greene, 2005 [ Reply ] By: Luca Elisei on 2016-08-24 09:12 | [forum:43467] |
|
Hi Arne, Thank you for your support. For “within” transformation I mean deviations from group means, adjustment inserting the within-group means of inputs in the main frontier model. Can the command that you have written, do this transformation? BR, Luca Elisei |
|
| RE: Stochastic frontier by true fixed effects (TFE) Greene, 2005 [ Reply ] By: Arne Henningsen on 2016-08-24 08:11 | [forum:43464] |
|
Dear Luca AFAIK, the true-fixed-effects model can be estimated by the following commands that are in the 'examples' section of the documentation of sfa(): library("frontier") data( "riceProdPhil" ) riceTrue <- sfa( log( PROD ) ~ log( AREA ) + log( LABOR ) + log( NPK ) + factor( FMERCODE ), data = riceProdPhil ) summary( riceTrue ) However, in contrast to Greene (2005), the optimization algorithm does not take into account the structure of the parameters and thus, is likely less efficient and may fail more frequently. What do you mean by "within-group means"? Best wishes, Arne |
|
| Stochastic frontier by true fixed effects (TFE) Greene, 2005 [ Reply ] By: Luca Elisei on 2016-08-16 13:47 | [forum:43429] |
|
How can I estimate SFM with true fixed effects (TFE) model of Greene, 2005 (within-group means) ? Thanks, Luca Elisei |
|
