# st: gllamm gllapred conditional effect plot?

 From [email protected] To [email protected] Subject st: gllamm gllapred conditional effect plot? Date Mon, 24 Sep 2007 13:37:39 +0200

Greetings, Statlisters:

does anyone know how to obtain a conditional effect plot for an
interaction effect in a 3-level gllamm model? Is there a way to
graphically show the level 1 interaciton effect just as in standard
logistic regression models? I would like to have the dependent var on
the y-achses and one interaction term on the x-achses, while plotting
two lines for the min and max value of the other interaction term. How
is this possible while holding covariates constant at their means (or
specified values)?

My model looks like this:

number of level 1 units = 4288
number of level 2 units = 984
number of level 3 units = 79

Condition Number = 24.125823

gllamm model

log likelihood = -1015.1815

------------------------------------------------------------------------------
emiprown2 | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
imp2 | -5.562665 .7225155 -7.70 0.000 -6.978769 -4.14656
opp2 | -3.925741 .9753384 -4.03 0.000 -5.837369 -2.014113
imp2_opp2 | 5.789333 1.817443 3.19 0.001 2.227211 9.351455
lev2 cov. | 1.261853 .576879 2.19 0.029 .1311909 2.392515
. | .6489876 .5654198 1.15 0.251 -.4592148 1.75719
. | .7975074 .8173145 0.98 0.329 -.8043996 2.399414
. | 1.383101 .7998135 1.73 0.084 -.1845043 2.950707
. | 2.011012 1.303253 1.54 0.123 -.5433177 4.565342
. | -.0540618 .4178598 -0.13 0.897 -.8730519 .7649283
. | -.1234842 .3242951 -0.38 0.703 -.7590909 .5121226
. | -.0351176 .5613223 -0.06 0.950 -1.135289 1.065054
. | -.4662562 .4107174 -1.14 0.256 -1.271248 .3387352
. | -.8884232 2.092618 -0.42 0.671 -4.98988 3.213034
. | 1.020689 .7510611 1.36 0.174 -.4513635 2.492742
lev3 cov | .1920239 .7266132 0.26 0.792 -1.232112 1.61616
. | -.3062174 1.015623 -0.30 0.763 -2.296803 1.684368
_cons | -2.801371 1.023168 -2.74 0.006 -4.806744 -.7959987
------------------------------------------------------------------------------

Variances and covariances of random effects
------------------------------------------------------------------------------

***level 2 (id)

var(1): 7.5339368 (1.2047601)

***level 3 (intno)

var(1): 1.1835858 (.48065168)
------------------------------------------------------------------------------

Thanks a lot,

Martin
-----------------
Martin Neugebauer
MA Student
University of Mannheim

*
* 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/