Re: st: create data for frontier estimation

[Federico Belotti <f.belotti@gmail.com>]

clear all

set obs 10000
set seed 543210
gen x1 = rnormal(2)
gen x2 = rnormal(1)
gen x3 = rnormal(1)

gen lnsig2u = - 1.2 + 1.4 * x3 

gen sigu = sqrt(exp(lnsig2u))
scalar sigv = 0.5 
gen u = abs(rnormal(0, sigu))
gen v = rnormal(0,sigv)

gen y = 0.5 * x1 + 0.5 * x2 + 0.5 + v + u

summ, sep(0)
frontier y x1 x2, cost uhet(x3)


On Oct 11, 2013, at 9:14 AM, Cyrus Levy wrote:

> I'm trying to create a dataset using the true model specified below. The outcomes look good except the constant in lnsig2u. I am setting it to -1.2 in the program, but the estimation result is -0.197. I'm not sure what's wrong with the program. How can I fix this problem?
> 
> drop _all
> set obs 10000
> set seed 543210
> gen x1 = rnormal(2)
> gen x2 = rnormal(1)
> gen x3 = rnormal(1)
> 
> gen lnsig2u = 1.4 * x3 - 1.2
> 
> gen sigu = sqrt(exp(lnsig2u))
> gen nsig = sqrt(sigu^2 / (1 - 2/_pi))
> gen u = abs(rnormal(0, nsig))
> gen v = rnormal()
> 
> gen y = 0.5 * x1 + 0.5 * x2 + 0.5 + v + u
> 
> summ, sep(0)
> frontier y x1 x2, cost uhet(x3)
> 
> 
> 
> Stoc. frontier normal/half-normal model           Number of obs =      10000
>                                                  Wald chi2(2) =    2656.28
> Log likelihood = -18811.039                       Prob > chi2 =     0.0000
> 
> ------------------------------------------------------------------------------
>           y |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> y            |
>          x1 |    .515271   .0137709    37.42   0.000 .4882805    .5422616
>          x2 |   .4967352   .0137501    36.13   0.000 .4697854    .5236849
>       _cons |   .4736053   .0428721    11.05   0.000 .3895775     .557633
> -------------+----------------------------------------------------------------
> lnsig2v      |
>       _cons |  -.0455195     .03432    -1.33   0.185 -.1127855    .0217464
> -------------+----------------------------------------------------------------
> lnsig2u      |
>          x3 |   1.407146   .0321241    43.80   0.000 1.344184    1.470108
>       _cons |  -.1970581   .0729905    -2.70   0.007    -.3401169 -.0539994
> -------------+----------------------------------------------------------------
>     sigma_v |   .9774973   .0167738 .9451679    1.010933
> ------------------------------------------------------------------------------ 
> *
> *   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/

-- 
Federico Belotti, PhD
Research Fellow
Centre for Economics and International Studies
University of Rome Tor Vergata
tel/fax: +39 06 7259 5627
e-mail: federico.belotti@uniroma2.it
web: http://www.econometrics.it


*
*   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/