st: RE: AW: RE: partial()-option of xtivreg2

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Jessica,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Jessica �lschl�ger
> Sent: Friday, December 21, 2007 9:27 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: AW: RE: partial()-option of xtivreg2
> 
> Dear Mark,
> 
> thank you for your reply.
> 
> I must admit that I'm not that experienced in statistics yet 
> and thus far from suggesting anything to you.
> But from my point of view I would prefer an informative error 
> message by stata when using the fd option. 
> 
> I meant to do the partialling-out after the first-differencing.
> 
> Would it make any sense to do the partialling-out before 
> first-differencing?

Austin Nichols suggests that partialling-out after the first differences is perfectly logical, and on reflection I think he's probably right.  Programming it into -xtivreg2- will go on my to-do list.

In the meantime, you can do what you want by hand: just generate new variables, e.g.,

gen double d_gwgdpvdpc = d.gwgdpvdpc

and then use -ivreg2- with the -partial- option and the first-differenced variables to do the partialling-out.

Cheers,
Mark


> I'm afraid, I'm at a complete loss on this topic. 
> 
> Regards 
> 
> Jessica
> 
>  
> ----------------------------------------------------------------
> Dipl. oec. Jessica �lschl�ger
>  
> Forschungsstelle Europ�ische Integration Universit�t Hohenheim
> 
> -----Urspr�ngliche Nachricht-----
> Von: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von 
> Schaffer, Mark E
> Gesendet: Donnerstag, 20. Dezember 2007 14:01
> An: statalist@hsphsun2.harvard.edu
> Betreff: st: RE: partial()-option of xtivreg2
> 
> Jessica,
> 
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Jessica 
> > �lschl�ger
> > Sent: Thursday, December 20, 2007 11:56 AM
> > To: statalist@hsphsun2.harvard.edu
> > Subject: st: partial()-option of xtivreg2
> > 
> > Dear Statalisters,
> > 
> > I'm trying to fit following model using xtivreg2. (I'm using an 
> > unbalanced panel data set.)
> > 
> > xtivreg2 gwgdpvdpc  ln_openquota_oecd ln_govbaladj_oecd 
> > ln_unemployment
> > ln_lgovinvquota  outputgap l.totalpop   rintratelt syr   
> > termstrade  fdiin
> > t1100gdp t1200gdp t2000gdp t3000gdp t5000gdp  if year >=1990, fd
> > cluster(coid)
> > 
> > Doing this, stata gives me the warning:
> > 
> > Warning: estimated covariance matrix of moment conditions 
> not of full 
> > rank.
> >          standard errors and model tests should be interpreted with 
> > caution.
> > Possible causes:
> >          number of clusters insufficient to calculate robust 
> > covariance matrix
> >          singleton dummy variable (dummy with one 1 and N-1 
> 0s or vice
> > versa)
> > partial option may address problem.
> > 
> > My question is, how should I use the partial-option? I've 
> got a real 
> > problem with understanding the syntax correctly. I always 
> get the same 
> > error message, e.g. when I try:
> 
> I think the problem is a misleading message by xtivreg2 
> rather than a bug per se.  Partialling-out is appropriate for 
> fixed effects models, but I'm not sure what it would mean 
> when combined with first differencing.
> 
> I'm inclined to make -partial- incompatible with -fd- so that 
> combining them causes -xtivreg2- to exit with an informative 
> error.  What do you think?  Or should it mean that the 
> partialling-out is applied to the data BEFORE first-differencing?
> 
> --Mark
> (-xtivreg2- author)
> 
> > . xtivreg2 gwgdpvdpc  ln_openquota_oecd ln_govbaladj_oecd 
> > ln_unemployment
> > ln_lgovinvquota  outputgap l.totalpop   rintratelt
> > >  syr   termstrade  fdiin   t1100gdp t1200gdp t2000gdp 
> > t3000gdp t5000gdp
> > if year >=1990, fd cluster(coid) partial(ln_openqu
> > > ota_oecd ln_govbaladj_oecd  ln_unemployment ln_lgovinvquota
> > > outputgap)
> > 
> > Error: ln_openquota_oecd ln_govbaladj_oecd ln_unemployment 
> > ln_lgovinvquota outputgap listed in partial() but not in list of r
> > > egressors.
> > invalid syntax
> > 
> > Could anybody please give me a hint what I am doing wrong?
> > Any comments would be greatly appreciated.
> > 
> > Regards
> > 
> > Jessica
> >  
> > ----------------------------------------------------------------
> > Dipl. oec. Jessica �lschl�ger
> >  
> > Forschungsstelle Europ�ische Integration Universit�t Hohenheim
> > 
> > 
> > *
> > *   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/
> > 
> 
> *
> *   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/
> 
> 
> *
> *   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/
> 

*
*   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/