Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Significance of mixed-effect parameter - arithmetic right, meaning wrong
From
"Allan Reese (Cefas)" <[email protected]>
To
<[email protected]>
Subject
st: Significance of mixed-effect parameter - arithmetic right, meaning wrong
Date
Thu, 24 Feb 2011 14:43:11 -0000
Playing with xtmelogit, I note that this first output shows the
random-effect factor is very significant by the LR test. The confidence
interval for the random-effect SD excludes 0 (=no random effect) and
happens to include 1.
The second output, with option -estm-, generates a test of the log
standard deviation against zero, equivalent to testing the sd(_cons)
against a null value of 1. P>.8 indicates not significantly different.
This makes no sense to me and appears a pointless and misleading piece
of output (the P value).
. xtmelogit sitenA lwmean || spn: , bin(Natstation) nolog
Mixed-effects logistic regression Number of obs =
58
Binomial variable: Natstation
Group variable: spn Number of groups =
6
Obs per group: min =
4
avg =
9.7
max =
24
Integration points = 7 Wald chi2(1) =
239.09
Log likelihood = -128.63521 Prob > chi2 =
0.0000
------------------------------------------------------------------------
------
sitenA | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
lwmean | -2.049232 .1325283 -15.46 0.000 -2.308983
-1.789481
_cons | 3.343523 .4401414 7.60 0.000 2.480862
4.206185
------------------------------------------------------------------------
------
------------------------------------------------------------------------
------
Random-effects Parameters | Estimate Std. Err. [95% Conf.
Interval]
-----------------------------+------------------------------------------
------
spn: Identity |
sd(_cons) | .9247444 .328285 .4611524
1.854381 <<< That's ok
------------------------------------------------------------------------
------
LR test vs. logistic regression: chibar2(01) = 75.99 Prob>=chibar2 =
0.0000
. xtmelogit sitenA lwmean || spn: , bin(Natstation) nolog estm
Mixed-effects logistic regression Number of obs =
58
Binomial variable: Natstation
Group variable: spn Number of groups =
6
Obs per group: min =
4
avg =
9.7
max =
24
Integration points = 7 Wald chi2(1) =
239.09
Log likelihood = -128.63521 Prob > chi2 =
0.0000
------------------------------------------------------------------------
------
sitenA | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
eq1 |
lwmean | -2.049232 .1325283 -15.46 0.000 -2.308983
-1.789481
_cons | 3.343523 .4401414 7.60 0.000 2.480862
4.206185
-------------+----------------------------------------------------------
------
lns1_1_1 |
_cons | -.078238 .3550008 -0.22 0.826 -.7740268
.6175509 <<< p>.8?!
------------------------------------------------------------------------
------
Regards
Allan
R Allan Reese
Senior statistician, Cefas
The Nothe, Weymouth DT4 8UB
Tel: +44 (0)1305 206614 -direct
Fax: +44 (0)1305 206601
www.cefas.co.uk
***********************************************************************************
This email and any attachments are intended for the named recipient only. Its unauthorised use, distribution, disclosure, storage or copying is not permitted. If you have received it in error, please destroy all copies and notify the sender. In messages of a non-business nature, the views and opinions expressed are the author's own and do not necessarily reflect those of the organisation from which it is sent. All emails may be subject to monitoring.
***********************************************************************************
*
* 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/