----- Original Message -----
From: <[email protected]>
To: <[email protected]>
Sent: Tuesday, January 28, 2003 3:01 AM
Subject: st: Re: RE: problem with -xtpcse-
> What I am wondering about is that although OLS on mean-differenced data
> (LSDV - what I am doing with -xtpcse-) should give the same coefficients as
> -areg, absorb(i)-, it does not. The coefficients differ quite a lot, and the
> Rsq with -xtpcse- is only half that with -areg,absorb(i)-. (I know that
> there is a difference in the calculation of Rsq, just would not have thought
> it would be so large. Thus I would already be happy if somebody could say
> something about why the coefficients might differ.)
>
Are you de-meaning the data correctly?
Using -xtpcse- with dummy variables, or using -xtdata, fe- to transform the data
and then -xtpcse- produces the same coefficient estimates as -areg-.
. webuse grunfeld
. areg inv mv kst, ab(co)
Number of obs = 200
F( 2, 188) = 309.01
Prob > F = 0.0000
R-squared = 0.9441
Adj R-squared = 0.9408
Root MSE = 52.768
------------------------------------------------------------------------------
invest | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mvalue | .1101238 .0118567 9.29 0.000 .0867345 .1335131
kstock | .3100653 .0173545 17.87 0.000 .2758308 .3442999
_cons | -58.74393 12.45369 -4.72 0.000 -83.31086 -34.177
-------------+----------------------------------------------------------------
company | F(9, 188) = 49.177 0.000 (10 categories)
. xi i.com
i.company _Icompany_1-10 (naturally coded; _Icompany_1 omitted)
. xtpcse inv mva kst _I*
Linear regression, correlated panels corrected standard errors (PCSEs)
Group variable: company Number of obs = 200
Time variable: year Number of groups = 10
Panels: correlated (balanced) Obs per group: min = 20
Autocorrelation: no autocorrelation avg = 20
max = 20
Estimated covariances = 55 R-squared = 0.9441
Estimated autocorrelations = 0 Wald chi2(11) = 4460.13
Estimated coefficients = 12 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
| Panel-corrected
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mvalue | .1101238 .0175568 6.27 0.000 .0757132 .1445344
kstock | .3100653 .0245731 12.62 0.000 .261903 .3582277
_Icompany_2 | 172.2025 49.40775 3.49 0.000 75.36509 269.0399
_Icompany_3 | -165.2751 48.564 -3.40 0.001 -260.4588 -70.09144
_Icompany_4 | 42.4874 64.11886 0.66 0.508 -83.18325 168.158
_Icompany_5 | -44.32013 76.28904 -0.58 0.561 -193.8439 105.2036
_Icompany_6 | 47.13539 68.53321 0.69 0.492 -87.18723 181.458
_Icompany_7 | 3.743212 74.43241 0.05 0.960 -142.1416 149.6281
_Icompany_8 | 12.75103 65.16085 0.20 0.845 -114.9619 140.464
_Icompany_9 | -16.92558 71.53431 -0.24 0.813 -157.1303 123.2791
_Icompany_10 | 63.72884 73.52344 0.87 0.386 -80.37446 207.8321
_cons | -70.29669 74.7058 -0.94 0.347 -216.7174 76.12399
------------------------------------------------------------------------------
. xtdata, fe
. xtpcse inv mv kst
Linear regression, correlated panels corrected standard errors (PCSEs)
Group variable: company Number of obs = 200
Time variable: year Number of groups = 10
Panels: correlated (balanced) Obs per group: min = 20
Autocorrelation: no autocorrelation avg = 20
max = 20
Estimated covariances = 55 R-squared = 0.7668
Estimated autocorrelations = 0 Wald chi2(2) = 299.86
Estimated coefficients = 3 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
| Panel-corrected
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mvalue | .1101238 .0175568 6.27 0.000 .0757132 .1445344
kstock | .3100653 .0245731 12.62 0.000 .261903 .3582277
_cons | -58.74393 18.00248 -3.26 0.001 -94.02813 -23.45973
------------------------------------------------------------------------------
.
Scott
*
* 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/
