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: plotting 95% trajectory curves
From
Thomas Norris <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: plotting 95% trajectory curves
Date
Wed, 23 Jan 2013 12:06:12 +0000
Dear statalisters,
I have ran a cubic polynomial multilevel model (weight is on the log scale) on a prenatal dataset with the command:
'xtmixed lnweight age age2 age3|| studyid: age age2,cov(unstructured)mle'. Output below:
------------------------------------------------------------------------------
lnweight | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------
-------------+------
age | .301346 .0094328 31.95 0.000 .2828581 .3198339
age2 | -.0025259 .0003151 -8.02 0.000 -.0031436 -.0019083
age3 | -.0000121 3.40e-06 -3.56 0.000 -.0000188 -5.45e-06
_cons | .8590841 .090142 9.53 0.000 .682409 1.035759
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------
-----------------------------+------
studyid: Unstructured|
sd(age) | .0187369 .0022426 .0148191 .0236905
sd(age2) | .0003173 .0000369 .0002526 .0003985
sd(_cons) | .2331245 .0350259 .1736595 .3129515
corr(age,age2) | -.977704 .005079 -.9857472 -.9652014
corr(age,_cons) | -.9588043 .0094474 -.9737672 -.9355854
corr(age2,_cons) | .9000741 .0237249 .8419113 .9375634
-----------------------------+------------------------------------------
-----------------------------+------
sd(Residual) | .0594447 .0008272 .0578452 .0610884
LR test vs. linear regression: chi2(6) = 3652.95 Prob > chi2 = 0.0000
I have predicted the mean curve for the sample, but would like to plot the 95% curves to visualise the variability in the curves. My random effects covariances are below.
age age2 _cons
age .0003511
age2 -5.81e-06 1.01e-07
_cons -.0041881 .0000666 .054347
------------------------------------------------------------------------------
Would anyone be able to help provide some code to help me fit these 95% curves?
Many thanks,
Tom
Tom Norris (PhD student)
Centre of Global Health and Human Development
School of Sport, Exercise and Health Sciences
Loughborough University
Loughborough
LE11 3TU
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/