Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: gmm and a panel data set


From   "Brian P. Poi" <bpoi@stata.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: gmm and a panel data set
Date   Thu, 23 Feb 2012 16:27:54 -0600


Marder, Andrew wrote:
Dear Statalist,

I am having some difficulty using the gmm command.  When I use the command as follows:

gmm (`moment_condition'), winitial(identity) instruments(L.(`instruments'))

The command is able to return some estimates.  I am unsure how valid these estimates are because when I run the command treating my data as a panel:

xtset firm_id year
gmm (`moment_condition'), winitial(xt L) xtinstruments(`instruments', lags(1))

I get the following error:

warning: 20398 missing values returned for equation 1 at initial values
             gmm_xtZ_i():  3301  subscript invalid
           gmm_ABond_W():     -  function returned error
              _gmm_wrk():     -  function returned error
                 <istmt>:     -  function returned error
r(3301);

From what I've read this could be due to an unbalanced panel or missing data.  Taking a look at my data, it is both unbalanced and has missing data:

. tsset
       panel variable:  firm_id (unbalanced)
        time variable:  year, 1989 to 2009, but with gaps
                delta:  1 unit

Do you have any knowledge of what is causing this error?  Any suggestions on how to estimate a GMM model using a messy panel data set?

Regardless of how your data is -tsset-, you should not get those cryptic error messages from Mata.  That appears to be a bug in -gmm-.  If you send your dataset and do-file to me privately, I'd be happy to look into the problem and report back to statalist once I figure out what is causing it.


    -- Brian Poi
    -- bpoi@stata.com


*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index