Statalist


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

st: disagreement between xtreg and xtmixed outputs


From   José Maria Pacheco de Souza <jmpsouza@usp.br>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: disagreement between xtreg and xtmixed outputs
Date   Mon, 29 Jun 2009 12:56:40 -0300

Dear  statalisters:
I ran xtreg and xtmixed (and 1oneway) with the same dataset but got different results, as I show bellow:

xtreg  mensuração , i(id) mle
Iteration 0:   log likelihood =  -38.39068
Random-effects ML regression Number of obs = 60 Group variable: id Number of groups = 30 Random effects u_i ~ Gaussian Obs per group: min = 2 avg = 2.0 max = 2 Wald chi2(0) = 0.00 Log likelihood = -38.39068 Prob > chi2 = .

mensuração Coef. Std. Err. z P>z [95% Conf. Interval]
_cons          2.132833   .1942075    10.98   0.000     1.752194    2.513473

/sigma_u 1.052955 .1387291 .8133212 1.363193 /sigma_e .2134579 0 .2134579 .2134579
rho    .9605257          .                             .           .
Likelihood-ratio test of sigma_u=0: chibar2(01)= 101.09 Prob>=chibar2 = 0.000


xtmixed  mensuração ||   id:, mle
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0:   log likelihood = -23.147063
Iteration 1:   log likelihood = -23.147063

Computing standard errors:
Mixed-effects ML regression Number of obs = 60 Group variable: id Number of groups = 30 Obs per group: min = 2 avg = 2.0 max = 2

Wald chi2(0) = . Log likelihood = -23.147063 Prob > chi2 = .


mensuração Coef. Std. Err. z P>z [95% Conf. Interval] _cons .132833 .1942075 10.98 0.000 1.752194 2.513473

Random-effects Parameters     Estimate   Std. Err.     [95% Conf. Interval]
id: Identity
          sd(_cons)    1.062051   .1375417       .823967    1.368929

          sd(Residual)    .0841923   .0108692      .0653706    .1084331
LR test vs. linear regression: chibar2(01) = 131.58 Prob >= chibar2 = 0.0000


. loneway  mensuração id
One-way Analysis of Variance for mensuração:
Number of obs =        60
R-squared =    0.9969

Source SS df MS F Prob > F

Between id             67.889766        29    2.3410264    330.26     0.0000
Within id                    .21265011     30    .00708834
Total                     68.102416          59    1.1542782

Intraclass       Asy.
correlation      S.E.       [95% Conf. Interval]
------------------------------------------------
0.99396     0.00222       0.98962     0.99831

Estimated SD of id effect               1.080263
Estimated SD within id                  .0841923
Est. reliability of a id mean            0.99697
(evaluated at n=2.00)

. dis sqrt(2.3410264)
1.5300413

. dis sqrt(.00708834)
.08419228

. dis 1.062051^2/((1.062051^2)+(.0841923^2))
.99375499

. dis 1.052955^2/(( 1.052955^2)+(.2134579^2))
.96052575

I am using version 10 updated, intercooled. The data is in the long format.
I followed the models presented in pages 64 and 65 of Sophia Rabe-Hesketh´s Multilevel and Longitudinal Modeling Using Stata, second edition and really am puzlled with the no concordance of results. Could it be some problem with my data set (which is not that presented by Sophia)?
Thank you for any advice.
José Maria

Jose Maria Pacheco de Souza, Professor Titular (aposentado)
Departamento de Epidemiologia/Faculdade de Saude Publica, USP
Av. Dr. Arnaldo, 715
01246-904  -  S. Paulo/SP - Brasil
fones (11)3061-7747; (11)3768-8612;(11)3714-2403
www.fsp.usp.br/~jmpsouza
*
*   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