Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

RE: st: year variables in xtabond


From   André Ferreira Coelho <andre.f.coelho2011@novasbe.pt>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: year variables in xtabond
Date   Thu, 03 Jan 2013 12:20:31 +0100

Dear Sjoerd,

Thank you for your answer and for the suggested paper, which I'll find
certainly helpful. 
Meanwhile I found that while I had generated missing values and only 
replaced those observations by 1, the example using abdata.dta has all 
the remaining observations replaced by zeros. I don't know why -xtabond- 
does not work in the way I had previously generated dummies..
Yes, you are right about including dummies and a year variable. I just
wanted to control for some more recent years in my database.

Appreciated for you help,

Andre





> From: vanbekkum@ese.eur.nl
> Date: Thu, 3 Jan 2013 03:03:06 +0100
> Subject: Re: st: year variables in xtabond
> To: statalist@hsphsun2.harvard.edu
> 
> Andre,
> 
> - Your "no observations" error can have many causes unrelated to
> estimation (such as including a string variable in your model).
> - The abond.dta contains dummies generated perhaps by something as
> "tab year, gen()". However, in your own dataset, abond probably works
> fine with a prefix such as "xi: abond y x .... i.year, ....."
> - I don't think you want to include the year variable AND the dummies
> - To use dynamic panel estimators wisely, read Roodman's paper
> available at http://www.cgdev.org/content/publications/detail/14256.and
> his other paper cited therein.
> 
> ---
> 
> Sjoerd van Bekkum
> Erasmus University, Netherlands
> W http://people.few.eur.nl/vanbekkum/ P (+31)-10-4082114 F (+31)
> 10 - 40 89165
> 
> 
> 
> On 2 January 2013 22:36, André Ferreira Coelho
> <andre.f.coelho2011@novasbe.pt> wrote:
> >
> > Dear statalist,
> >
> > I have a basic question regarding xtabond. I have just realized that
> > factor variables are not allowed when using -xtabond-.
> > How would you control then for individual year effects? I simply tried
to
> > generate dummy variables for each year but seems that -xtabond-routine
does
> > not handle them since an error message stating that there are no
> > observations is issued.
> > Stata manual uses abdata.dta,-webuse abdata-, which contains dummy
> > variables in such a way they are labelled for instance as
"year==1980",
> > which means that they were generated considering in some way the time
> > variable year applied in xtset.
> >
> > So, basically what I pretend is just to include individual time
controls
> > such as those used in the manual (i.e. yr1980-yr1984) in addition to
the
> > variable year.
> >
> > Do you know how to do this?
> >
> > Best regards,
> >
> > Andre Coelho
> > *
> > * 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/
> 
> *
> * 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/

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index