Re: st: xtreg or xtmixed?

[Johan van der Watt <jjvanderwatt@gmail.com>]

Thank you so much!
I created dummy variables for visit (visit 1/baseline as ref), then
created variables for the interaction with neuropathy:
gen neu_vis2= neuropathy*visit2
gen neu_vis3=neuropathy*visit3

I then added them to my -xtreg- syntax:
.xtreg cd4_ neuropathy visit neu_vis2 neu_vis3, i(patientno) mle

This is my output. To interpret this I say there is no difference in
CD4 over time between those with neuropathy vs those without
(p=0.402), neither at visit 2 or visit 3 (p=0.481; p=0.904), however
there is a significant difference in CD4 over time looking at the
complete cohort (p=0.000).
                 cd4_ |                      Coef.   Std. Err.      z
  P>|z|     [95% Conf. Interval]
----------------------+----------------------------------------------------------------
        neuropathy         |  -.1288783   .1539001    -0.84   0.402
-.4305168    .1727603
                visit           |   .1379031   .0130418    10.57
0.000     .1123417    .1634646
             neu_vis2       |  -.0865857   .1227696    -0.71   0.481
 -.3272096    .1540382
             neu_vis3       |  -.0157075   .1306693    -0.12   0.904
 -.2718145    .2403996
                _cons        |   3.804473   .0809168    47.02   0.000
   3.645879    3.963067
----------------------+----------------------------------------------------------------
             /sigma_u      |   .6622547   .0466019
      .5769355    .7601912
             /sigma_e      |   .4882691    .020833
       .4490978     .530857
                  rho          |   .6478421   .0393632
          .5681531    .7215083


Thanks again!
Johan






                 cd4_ |      Coef.   Std. Err.      z    P>|z|
[95% Conf. Interval]
----------------------+----------------------------------------------------------------
totalincidworse_sympt |  -.1445857    .152609    -0.95   0.343
-.4436939    .1545224
visit |                             .1379031   .0130418    10.57
0.000     .1123417    .1634646
neu_vis1 |                      .0157075   .1306693     0.12   0.904
 -.2403996    .2718145
neu_vis2 |                      -.0708782   .1139779    -0.62   0.534
  -.2942709    .1525145
neu_vis3 |          0  (omitted)
_cons |   3.804473   .0809168    47.02   0.000     3.645879    3.963067
----------------------+----------------------------------------------------------------
             /sigma_u |   .6622547   .0466019
.5769355    .7601912
             /sigma_e |   .4882691    .020833
.4490978     .530857
                  rho |   .6478421   .0393632
.5681531    .7215083


On Mon, Dec 19, 2011 at 2:18 PM, Maarten Buis <maartenlbuis@gmail.com> wrote:
> On Mon, Dec 19, 2011 at 12:39 PM, Johan van der Watt wrote:
>> The question that I would like to answer, is whether the course of CD4
>> count over time is different in the group of patients who have
>> neuropathy vs the group who do not have neuropathy. Firstly, my
>> dependent variable is binary, either yes or no for development.
>
> If that is your question than neuropathy is _not_ your
> dependent/explained/left-hand-side/y variable but your
> independent/explanatory/right-hand-side/x variable. The dependent
> variable is the thing you want to explain, in this case CD4 count. The
> independent variable is what you use to do the explaining, in this
> case neuropathy.
>
>> Secondly, I am struggling with the fact that some of the continuous
>> variables are non-linear when I plot them over time, in other words,
>> as an example, I find inflammatory markers going up from baseline
>> between the 2nd and 3 visit, and then going back to baseline level at
>> the last visit. This pattern is the same in both groups (neuropathy
>> and neuropathy free), however I suspect the magnitude might be
>> different.
>
> That sounds like you want to add dummies for three of your time
> periods (treat baseline as the reference) and add interactions between
> neuropathy and the time dummies.
>
>> Thank you for the slides, I am just not sure how to deal
>> with this in longitudinal data.
>
> The point you need to take from that is not expect that the
> distributions of your dependent variables are normally distributed.
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
>
> http://www.maartenbuis.nl
> --------------------------
> *
> *   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/

*
*   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/