Dear Stata-users,
My question is not related to Stata, but to general econometrics. I have
an interpretational problem regarding coefficients of a Translog
production function. The model is to see an effect of a regulatory
variable on GDP:
lnY = b + bK*lnK + bL*lnL + bKK*(lnK)^2 + bLL*(lnL)^2
+ bR*lnR + bRK*(lnR)*(lnK) + bRL*(lnR)*(lnL)
where R is for regulatory proxy and b's are for coefficient estimates.
Q1) The data are divided into geographic groups. For most groups, bK and
bLL got negative signs whereas bKK and bL are positive. Do the
coefficients for capital--convexity w.r.t capital-- make an economic
sense?
Q2) Suppose that the above model as a whole gains a statistical
significance over a restricted Cobb-Douglas model through a chi-square
test, while some of variables are not significant.
To calculate marginal effects of lnL, I use the first derivative of lnY
w.r.t. lnL, that is, bL + 2*bLL*(lnL) + bRL*(lnR).
If bL is not significant, then should I drop or keep bL to calculate the
marginal effect of lnL on lnY?
Any comments are welcomed.
Sun H.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
