st: re: creating constraints in panel data

[Kit Baum <baum@bc.edu>]

<>
I hope you can help me. I am trying to estimate a cobb-douglas function
(Y=X1^a*X2^b*X3^c*....) using panel data with fixed effects.

I am using the log of the production function as my functional form,
that is, Ln(Y) = aLn(X1) + bLn(X2) + cLn(X3)+....

And I need to assume that the sum of the exponentials must be equal to 1
(a+b+c+...=1)

Can I make such constraint for panel data models? I know I can use
"cnsreg" command, but this is not panel data estimation, and it is not
valid in may case.

It is always possible to impose that constraint on any linear regression by just substituting it in algebraically.
I.e.

(lnY - lnX3) = a (lnX1-lnX3) + b(lnX2-lnX3)

lincom can compute the missing c coefficient.

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


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