Martin,
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Martin Weiss
> Sent: 10 March 2009 19:24
> To: [email protected]
> 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" <[email protected]>
> To: <[email protected]>
> 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: [email protected]
> >> [mailto:[email protected]] On Behalf Of
> >> Schaffer, Mark E
> >> Sent: Tuesday, March 10, 2009 6:34 PM
> >> To: [email protected]
> >> 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: [email protected]
> >> > [mailto:[email protected]] On Behalf Of
> >> > Davide Cantoni
> >> > Sent: Tuesday, March 10, 2009 6:13 PM
> >> > To: [email protected]
> >> > 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: [email protected]
> >> > > [mailto:[email protected]] Im Auftrag
> >> > von Davide Cantoni
> >> > > Gesendet: Dienstag, 10. März 2009 16:29
> >> > > An: [email protected]
> >> > > 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/
