Mark,
Appended you will find a copy of our output.
I appreciate all your help.
Rafa
xtregar polity2 PPPopenness , re
RE GLS regression with AR(1) disturbances Number of obs =
702
Group variable (i): cty1 Number of groups =
37
R-sq: within = 0.0030 Obs per group: min =
2
between = 0.5743 avg =
19.0
overall = 0.3051 max =
24
Wald chi2(2) =
1.57
corr(u_i, Xb) = 0 (assumed) Prob > chi2 =
0.4558
------------------- theta --------------------
min 5% median 95% max
0.5547 0.5952 0.6532 0.6577 0.6577
----------------------------------------------------------------------------
--
polity2 | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
PPPopenness | .0136209 .0108653 1.25 0.210 -.0076747
0349165
_cons | 1.957344 .980394 2.00 0.046 .0358073
3.878881
-------------+--------------------------------------------------------------
--
rho_ar | .84057188 (estimated autocorrelation coefficient)
sigma_u | 5.4429879
sigma_e | 1.8583632
rho_fov | .89559999 (fraction of variance due to u_i)
----------------------------------------------------------------------------
--
xtregar polity2 PPPopenness , fe
FE (within) regression with AR(1) disturbances Number of obs =
665
Group variable (i): cty1 Number of groups =
37
R-sq: within = 0.0003 Obs per group: min =
1
between = 0.5348 avg =
18.0
overall = 0.3204 max =
23
F(1,627) =
0.22
corr(u_i, Xb) = 0.6142 Prob > F =
0.6407
----------------------------------------------------------------------------
--
polity2 | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
PPPopenness | .0053465 .0114505 0.47 0.641 -.0171395
0278326
_cons | 2.181906 .0781426 27.92 0.000 2.028453
2.335359
-------------+--------------------------------------------------------------
--
rho_ar | .84057188
sigma_u | 6.6058579
sigma_e | 1.8267237
rho_fov | .92896283 (fraction of variance due to u_i)
----------------------------------------------------------------------------
--
F test that all u_i=0: F(36,627) = 6.67 Prob > F =
0.0000
. which xtregar
I:\W5\STATA\ado\updates\x\xtregar.ado
*! version 1.2.13 20jun2003
----- Original Message -----
From: "Mark Schaffer" <[email protected]>
To: <[email protected]>
Sent: Wednesday, May 19, 2004 4:37 PM
Subject: Re: st: Fixed effects and e(sample)
> Rafa,
>
> From: "R.E. De Hoyos" <[email protected]>
> To: <[email protected]>
> Subject: st: Fixed effects and e(sample)
> Date sent: Wed, 19 May 2004 16:13:31 +0100
> Send reply to: [email protected]
>
> > We are using panel data and for some reason when I run fixed effects or
> > random effects the sample size is different. My hypothesis here is that
> > observations with one observation are dropped automatically but I
haven't
> > thought about this enough.
>
> I tried to replicate the phenomenon by creating a dataset in which
> one group has one observation, and failed. Both re and fe used the
> same number of observations.
>
> Can you post the output of the two commands, if it isn't too bulky?
> (If the output is voluminous, you can send it to me privately.) Also
> check which version of xtreg you're using (-which xtreg-).
>
> --Mark
>
> > However stata seems to only tell me it's dropping observations in the
> > observation count when I run fixed effects but is actually keeping the
> > same number of observations in e(sample) as the random effects has.
> >
> > Is there a problem in using e(sample) when using fixed effects? If so
what
> > variable stores the values?
> >
> > Thanks,
> >
> > Rafa
> > *
> > * 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/
>
>
> Prof. Mark E. Schaffer
> Director
> Centre for Economic Reform and Transformation
> Department of Economics
> School of Management & Languages
> Heriot-Watt University, Edinburgh EH14 4AS UK
> 44-131-451-3494 direct
> 44-131-451-3008 fax
> 44-131-451-3485 CERT administrator
> http://www.som.hw.ac.uk/cert
> *
> * 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/
