sf_transform STATA 10.1

[Sara Restrepo]

Good night,

I am working with stochastic frontier production function with the Kumbhakar book. The book uses Stata 10.1 version. It has command called "sf_transform" and It is not recognized by newer versions of Stata.
The command recovers the estimates of the variance of the errors (u and v (stochastic and deterministic errors) . The thing is that, the variance parameters are parameterized as exponential function. And the command automitized the process and continues to report the standard error, z statistic, p value, and the confidence interval.

I was wondering if someone knows the command in stata 14, or how to reply the process automated by this command.

Thanks!

[William Lisowski]

Welcome to Statalist!

It is not clear from your post what you have done. Specifically, have you followed the guidance in Appendix E on installing their user-written software from https://sites.google.com/site/sfbook2014/ ?

[Richard Williams]

I don't have the book. But there is a section entitled

E.1 Download and Install the User-Written Programs

I would check it out.

[Richard Williams]

I am reviving this ancient thread because somebody asked me the same question today. Instructions for installing the commands in Stata 12+ are at

https://sites.google.com/site/sfbook...ta-v12-v13-v14

I gave the commands

net install sfbook_install, from(https://sites.google.com/site/sfbook2014/home/install/) replace
sfbook_install

and it seemed to work.

[tig som]

Richard Williams I am currently working on utilizing your “sfprim” STATA code by Kumbhakar, Wang and Horncastle (2015) to estimate output-oriented allocative and technical inefficiencies. I found this command after installing the program of the book following your above instruction code. I noticed that the “sfprim” command provides options for cost and profit, but not for revenue. However, I am facing a challenge as I have data and variables for the revenue function but not for cost or profit. Could you please offer me any guidance on how I can use” sfprim” for my revenue function?

To provide some context, I have three products, each with its respective price. I have information on total revenue, revenue share for each product, and the amount for one fixed input. Thus, I am interested in allocative inefficiency in terms of suboptimal output mix and technical inefficiency in terms of suboptimal output level. My initial approach was to employ the shadow price method outlined in Chapter 6 of Kumbhakar and Lovell (2003). I planned to use a translog revenue function and the nonlinear iterative generalized least square method. However, since I am unable to find sufficient guiding materials in this area, I decided to explore the primal approach to help me decompose between technical and allocative inefficiency.

My challenge now is that the primal approach command only offers options for cost or profit, not revenue. Do you have any tips on adapting a primal approach for revenue function? I’m considering using a negative revenue as the "Y"(dependent ) variable and running “sfprim” with the cost option. Would this be a viable solution?

Thank you, and I hope to receive your tips!

[Richard Williams]

Hi Tig. I'm afraid I know nothing about the programs, I just saw the instructions for installing them. You may want to start a new thread with a more precise title. Hopefully you'll attract the attention of people who do know.

[tig som]

Thank you very much for your reply!! Best regards!!

[Johnstone kumwenda]

hello prof, the command you gave aint working in my stata 17: net install sfbook_install, from(https://sites.google.com/site/sfbook2014/home/install/) replace sfbook_install