Forum: help

RE: Getting started with R [ Reply ]
By: Kidu Hailu on 2013-11-03 15:29

Dear Arne,

Thank you again for suggesting me these useful materials to work with Package frontier.

I am now trying to estimate a production function by using the command sfa(), applying the model of Battese and Coelli (1995) in order to see the determinants of the estimated inefficiency.

I specified the following translog function:
translogProd <- (y ~ x1 + x2 + x3 + time + halfx1sq + halfx2sq + halfx3sq + tsq + x1x2 + x1x3 +x1t +x2x3 + x2t + x3t| z1 + z2 +z3 +z4 + z5 +z6 + z7 + z8 + time, data=sfadata)

However, the program reports the following warning messages:

Warning messages:
1: the parameter 'gamma' is close to the boundary of the parameter space [0,1]: this can cause convergence problems and can negatively affect the validity and reliability of statistical tests and might be caused by model misspecification
2: the covariance matrix of the maximum likelihood estimates is singular

Could you please suggest how to solve this problem? I attach the data herewith.
In addition, would like to estimate model by certain categories (such as province). How could I use "IF" condition (to say if province=1) in the command?

Thank you very much!

With regards,
Kidu

RE: Getting started with R [ Reply ] By: Arne Henningsen on 2013-10-21 06:03	[forum:39915]
You are welcome! Yes, please do not hesitate to ask further questions about frontier through this forum. It will be easiest to help you, if you provide a reproducible example. Cheers, Arne.

RE: Getting started with R [ Reply ]
By: Kidu Hailu on 2013-10-21 03:54

[forum:39914]

Dear Arne,

Thank you very much for the timely response to my query. The materials you suggested me are really helpful. As I read their difference, the R package "frontier" seems to be preferable to FRONTIER 4.1. Let me read more based on the materials you suggested me and I hope you will not mind if I coma back for further questions if I encounter difficulties as to how to apply the the R package "frontier".

Thank you for your understanding.

With best regards,
Kidu

RE: Getting started with R [ Reply ]
By: Arne Henningsen on 2013-10-20 20:28

[forum:39913]

Dear Kidu

I suggest that you use the R package "frontier" rather than running "FRONTIER 4.1" from within R. The differences between "FRONTIER 4.1" and the R package "frontier" are described at frontier's project home page at R-Forge [1]. The R-Project has several official manuals [2], e.g. "An Introduction to R", which is nice introduction to beginners. Furthermore, there are many "unofficial" R manuals freely available, e.g. several English and Non-English manuals that are listed at [3]. Many R users (from beginners to advanced developers) like to run R through RStudio IDE [4], which makes R (in my opinion) more user-friendly. If you have learned the basics of R, you might find it helpful to take a look at my collected lecture notes "Introduction to Econometric Production Analysis with R" [5]. Finally, you should carefully read the documentation of the frontier package [6]. And please do not forget to cite the "frontier" package in your publications (see output of the command 'citation("frontier")').

[1] http://frontier.r-forge.r-project.org/

[2] http://cran.r-project.org/manuals.html

[3] http://cran.r-project.org/other-docs.html

[4] http://www.rstudio.com/ide/

[5] http://absalon.itslearning.com/data/ku/103018/teaching/lectureNotes.pdf

[6] http://cran.r-project.org/web/packages/frontier/frontier.pdf

Best regards,
Arne

Getting started with R [ Reply ]
By: Kidu Hailu on 2013-10-15 17:33

[forum:39903]

Dear Arne,

I am studying on the relationship between firm size, age and technical efficiency. I have an unbalanced panel data of 1851 firms (8303 observation). I would like to apply the Battase and Coelli (1995) model in R. However, I am new to R and getting confused as to how to start the program. I would greatly appreciate if you could suggest me how to quick start the program relating to SFA (how to install the FRONTIER 4.1 or whichever you think is preferable). I already have the R software in my computer.

I look forward to hearing from you.
Kidu