Statalist


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

st: RE: Re: RE: RE: Re: AW: AW: ereturn list with xtivreg2


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Re: RE: RE: Re: AW: AW: ereturn list with xtivreg2
Date   Tue, 10 Mar 2009 22:48:21 -0000

Martin,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Martin Weiss
> Sent: 10 March 2009 19:24
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Re: RE: RE: Re: AW: AW: ereturn list with xtivreg2
> 
> <>
> 
> Mark,
> 
> Now that you are with us, just briefly: Is there a reason for 
> e(df_r) not to 
> be there in case -small- is not specified, or is this an omission?

It's been ages since I programmed it, but as I recall it was deliberate - the notion of "residual degrees of freedom" seemed to me a small-sample notion.

--Mark

> 
> 
> HTH
> Martin
> _______________________
> ----- Original Message ----- 
> From: "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
> To: <statalist@hsphsun2.harvard.edu>
> Sent: Tuesday, March 10, 2009 8:12 PM
> Subject: st: RE: RE: Re: AW: AW: ereturn list with xtivreg2
> 
> 
> > David,
> >
> > Ignore my email below, at least with respect to e(df_r).  
> Martin found the 
> > answer: you need the -small- option.
> >
> > I think you still have a problem, though - too few clusters to 
> > use -cluster-.
> >
> > --Mark
> >
> >> -----Original Message-----
> >> From: owner-statalist@hsphsun2.harvard.edu
> >> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of
> >> Schaffer, Mark E
> >> Sent: Tuesday, March 10, 2009 6:34 PM
> >> To: statalist@hsphsun2.harvard.edu
> >> Subject: st: RE: Re: AW: AW: ereturn list with xtivreg2
> >>
> >> Davide,
> >>
> >> I'm the author of -xtivreg2- so maybe I should take over from
> >> Martin...
> >>
> >> Can you run your regression again WITHOUT cluster and send us
> >> the list of returned macros?  I think part of the problem is
> >> that you have more regressors than clusters, and the
> >> cluster-robust var-cov matrix is not full rank.
> >>
> >> Can you also tell us which version you are using?  -which
> >> xtivreg2- will do it.
> >>
> >> And can you share the command line and output with us?
> >>
> >> --Mark
> >>
> >> > -----Original Message-----
> >> > From: owner-statalist@hsphsun2.harvard.edu
> >> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of
> >> > Davide Cantoni
> >> > Sent: Tuesday, March 10, 2009 6:13 PM
> >> > To: statalist@hsphsun2.harvard.edu
> >> > Subject: st: Re: AW: AW: ereturn list with xtivreg2
> >> >
> >> > I get a gazillion of results, just no e(df_r). Cf:
> >> >
> >> > (And, as I said, I have the same issue whether or not I have
> >> > clustering or not, robust se or not, etc etc.)
> >> >
> >> > > scalars:
> >> > >          e(inexog_ct) =  11
> >> > >          e(exexog_ct) =  1
> >> > >           e(endog_ct) =  1
> >> > >         e(partial_ct) =  0
> >> > >            e(N_clust) =  8
> >> > >          e(sdofminus) =  0
> >> > >           e(dofminus) =  8
> >> > >                 e(r2) =  .9229592972559002
> >> > >               e(rmse) =  4.482742177668595
> >> > >                e(rss) =  1728.168059104612
> >> > >                e(mss) =  20703.71531616696
> >> > >               e(r2_a) =  .9021374857034408
> >> > >                  e(F) =  2.799026483928738
> >> > >                 e(Fp) =  .0897520696224036
> >> > >               e(Fdf1) =  12
> >> > >               e(Fdf2) =  7
> >> > >                 e(yy) =  22431.88337527158
> >> > >                e(yyc) =  22431.88337527158
> >> > >                  e(N) =  94
> >> > >               e(df_m) =  20
> >> > >          e(singleton) =  0
> >> > >            e(sigma_e) =  4.482742177668595
> >> > >               e(df_a) =  8
> >> > >               e(df_b) =  12
> >> > >                e(N_g) =  8
> >> > >              e(g_max) =  12
> >> > >              e(g_min) =  11
> >> > >              e(g_avg) =  11.75
> >> > >        e(partialcons) =  0
> >> > >               e(cons) =  0
> >> > >                e(cdf) =  25.73075120843825
> >> > >            e(widstat) =  8.208461212944977
> >> > >                 e(cd) =  .3477128541680844
> >> > >                e(idp) =  .2053493221619689
> >> > >               e(iddf) =  1
> >> > >             e(idstat) =  1.603914697363933
> >> > >                e(jdf) =  0
> >> > >                  e(j) =  .
> >> > >                 e(ll) =  -270.2217758044246
> >> > >              e(rankV) =  7
> >> > >              e(rankS) =  7
> >> > >             e(rankxx) =  12
> >> > >             e(rankzz) =  12
> >> > >                e(r2c) =  .9229592972559002
> >> > >                e(r2u) =  .9229592972559002
> >> >
> >> > Davide
> >> >
> >> >
> >> > > <>
> >> > >
> >> > >
> >> > > You are more than welcome :-) Your issue is weird indeed!
> >> > What do you get
> >> > > with -ereturn list- after applying -xtivreg2- to your data?
> >> > I mean, does
> >> > > Stata populate the rest of the returned values?
> >> > >
> >> > >
> >> > >
> >> > > HTH
> >> > > Martin
> >> > >
> >> > >
> >> > > -----Ursprüngliche Nachricht-----
> >> > > Von: owner-statalist@hsphsun2.harvard.edu
> >> > > [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag
> >> > von Davide Cantoni
> >> > > Gesendet: Dienstag, 10. März 2009 16:29
> >> > > An: statalist@hsphsun2.harvard.edu
> >> > > Betreff: st: Re: AW: AW: ereturn list with xtivreg2
> >> > >
> >> > > Well, curious... I have -adoupdate-d my -xtivreg2-, just
> >> to be sure.
> >> > > When I run your example, I get in fact e(df_r) (thanks for
> >> > pointing it
> >> > > out in red!). But when I run my own command on my
> >> dataset, I do not
> >> > > get anything:
> >> > >
> >> > > . disp e(df_r)
> >> > > .
> >> > >
> >> > > This happens with and without clustering, with and 
> without robust
> >> > > s.e., with and without excluded instruments. In every 
> one of these
> >> > > cases, I do not have anything saved in e(df_r). My data
> >> are properly
> >> > > -tsset-.
> >> > >
> >> > > Davide
> >> >
> >> > *
> >> > *   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/
> >> >
> >>
> >>
> >> -- 
> >> Heriot-Watt University is a Scottish charity
> >> registered under charity number SC000278.
> >>
> >>
> >> *
> >> *   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/
> >>
> >
> >
> > -- 
> > Heriot-Watt University is a Scottish charity
> > registered under charity number SC000278.
> >
> >
> > *
> > *   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/
> > 
> 
> *
> *   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/
> 


-- 
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   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   |   What's new   |   Site index