st: RE: frontier

[ogundarikolawole <ogundarikolawole@daad-alumni.de>]

This is my response to Khushid.

Khushid's request on efficiency is within the frontier command as earier mentioned by Nick. 

Back to Khushid, to estimate the cost function or profit function you need a strong background of production economic theory. This is because your estimates must be consistent with economic theory. based on this it cost function and profit function require that you impose some sort of restriction on the functional form and its necessay that you also estimate these functions using flexible funcional form like translog.
You need to transform the data first into logarithm and use one of the prices to impose mononicity restriction on the cost function. For the profit function you use the price of the output to impose this restriction.

Model: 

logTC = a+b1logPBF +b2logOV+b3 logN1+b4 logN2 +b5logTL +b6logNFC b7Y + V+U

command in STATA

frontier log TC  logPBF logOV logN1 logN2 logTL logNFC Y, uhet(r, g, inf ) cost

But you need to use one of the prices in the frontier function to impose the restriction. I do not know which of the variables is in price unit. Let say N1 is in unit price, then you can specified the function as

frontier log(TC/N1) logPBF logOV log(N2/N1) logTL logNFC Y, uhet(r, g, inf ) cost

If PBF and OV are traditional inputs and not in price unit, you cannot normalize these variables. Also you cannot mormalize ouputs. You can only normalize TC and the prices. In that case you have impose restriction on the frontier to ensure that you a well behave frontier. of course, you need to read production economic literature why this is essential. I cannot explain this on this medium.

Alternatively, you can estimate input distance function to avoid which is the same as cost fucntion.

The command is "frontier" and uhet(r,g,inf) is another way of accounting for soucrces of inefficiency in the cost frontier. The appended "cost" is to ensure that  you are estimating cost frontier not production frontier which is the default in STATA.

Kolawole





Dr. Kolawole OGUNDARI
JSPS Research Fellow
Laboratory of Agricultural and Farm Management,
Dept. of Agricultural and Resources Economics, Faculty of Agriculture,

Kyushu University, Hakozaki 6-10-1, Fukuoka, 812-8581, Japan.

________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Khurshid Djalilov [kdjalilov@bournemouth.ac.uk]
Sent: Monday, December 17, 2012 3:52 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: frontier

I am trying to investigate Cost (Profit) inefficiencies for 86 banks representing two different regions (panel data).
In addition to input-output variables I have dummy as well as economic development variables. I was wondering how I can generate cost and profit efficiencies as I am a new user and am having problems with this.

The variable I have as following:

TC - dependent variable, PBF - (input, independent variable), OV - (input, independent variable), TL - (output, independent variable), OEA - (output, independent variable), NFC - (output, independent variable), N1 - (netput, independent variable), N2 - (netput, independent variable),
Y - (time, independent), r - regional dummy (independent), g - growth (independent), inf - inflation (independent).

I would appreciate if anyone could write commands using these variables, please.
Thank you.


BU is a Disability Two Ticks Employer and has signed up to the Mindful Employer charter. Information about the accessibility of University buildings can be found on the BU DisabledGo webpages [ http://www.disabledgo.com/en/org/bournemouth-university ]
This email is intended only for the person to whom it is addressed and may contain confidential information. If you have received this email in error, please notify the sender and delete this email, which must not be copied, distributed or disclosed to any other person.
Any views or opinions presented are solely those of the author and do not necessarily represent those of Bournemouth University or its subsidiary companies. Nor can any contract be formed on behalf of the University or its subsidiary companies via email.




*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/