Tim R. Sass <[email protected]> encounted an error message from -xtabond-
that is confusing. Tim entered the command
> xtabond nrtrscre charter chrtageb
> target convrt forprft
> nschools chgschl
> tgrde_06 tgrde_07 tgrde_08 tgrde_09 tgrde_10
> if cohort !=0, robust;
which yielded the error message
> too many lags of the dependent variable requested
> lags() must be beween 1 and 0
> r(198);
Since, Tim did not specify the number of lags(), the error message is
confusing.
This error message is issued when the number of lags of the dependent
variable is too large for the specified sample. So the problem is that
there are not enough time-series observations to proceed with the
estimation.
I would suggest that Tim take a quick look at the sample he has after
imposing the condition -if cohort !=0-.
For instance, if the time variable is -t- and id variable is -id-, Tim might
find the command
. graph dot id if cohort !=0, over(t)
useful in identifying the gaps in his sample.
--David
[email protected]
*
* 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/
Tim R. Sass
