Dear StataList Members:
I can not find the answer to the problem of "Stata drops variables in the
dataset" in the on-line resources. Maybe it is too basic to you fellow
members. I find it is different from "Why do estimation commands sometimes
drop variables?" in http://www.stata.com/support/faqs/stat/drop.html.
I run a estimation command with some different options. After the first time
of running the command, Stata randomly drops variable in the dataset. I then
have to exit and re-enter to run the command again. An example is below. It
happens frequently.
Could any colleague kindly point out the place where I can find the remedy
measures?
Thanks for your abundant knowledge and experience in advance!
Can Huang
xtgls tfp rdper rdexp newprodin fixassetin, i(number) t(year)
Cross-sectional time-series FGLS regression
Coefficients: generalized least squares
Panels: homoskedastic
Correlation: no autocorrelation
Estimated covariances = 1 Number of obs =
57
Estimated autocorrelations = 0 Number of groups =
2
Estimated coefficients = 5 Obs per group: min =
28
avg =
28.5
max =
29
Wald chi2(4) =
0.94
Log likelihood = 81.86981 Prob > chi2 =
0.9192
----------------------------------------------------------------------------
--
tfp | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
rdper | .000026 .000034 0.76 0.445 -.0000407
0000927
rdexp | .0008324 .0013939 0.60 0.550 -.0018996
0035644
newprodin | -.0238213 .0744405 -0.32 0.749 -.1697219
1220793
fixassetin | .0001438 .0005998 0.24 0.811 -.0010317
0013193
_cons | -.0172437 .0113719 -1.52 0.129 -.0395321
0050447
----------------------------------------------------------------------------
--
xtgls tfp rdper rdexp newprodin fixassetin, i(number) t(year)
panels(heterosk
> edastic)
variable rdper not found
r(111);
exit
*
* 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/
