st: When does one use liv() in xtdpd

[Hewan Belay <hewan_belay@yahoo.com>]

Dear Statalisters,

I have a short and simple question: Under what conditions does one use
the liv() option in xtdpd (dynamic models)? The manual's only comment on that option is: 

"liv(varlist) specifies additional standard instruments for the level equation. Specified variables may not be included in iv() or in div(). Levels of the variables are used as instruments for the level
equation. You may specify as many additional sets of standard instruments for the level equation as you need within the standard Stata limits on matrix size."
and
"strictly exogenous variables are handled using div(), iv(), or liv()".

All empirical examples of dynamic models given in the manual (and elsewhere) always 
use the div() option for using exogenous variables as instruments--whether 
the estimation is of the difference-GMM or system-GMM type. liv() in 
contrast is never used. Can 
someone give me an example or a context in which one may want to use liv()? 

I am wondering whether the usefulness of liv() may come in when, in a 
system-GMM estimation, one wants to make use of time-invariant exogenous 
regressors as instruments--since these can't be included in div() as they 
obviously drop out of the instrument set when differenced. Is my hunch 
correct, or is liv() for other uses?

Thanks,
Hewan

*
*   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/