sas >> Question about %glimmix and/or proc glimmix

by serinachiu » Mon, 22 Sep 2008 15:12:27 GMT

Hi guys,

I would like to create a SAS code using either %glimmix or proc
glimmix (though I am more familiar with %glimmix)

here is how my data looks like
I have a list of variable names such as d01-d07

and I would like to have all the variables under model statement with
intercept and one variable at a time under random statement... in this
case, which means, I will have to have 7 separate glimmix runs

I've tried to create a SAS macro with %do i=1 %to 7 with %glimmix,
however it only gave me the result on i=1, is there anyone could help
me out on this? thanks! or maybe gives me some idea on how to program
a DO LOOP with glimmix procedure! thanks again!

here is my %glimmix program:

%macro test(n=7,);
%Do i=1 %to &n;
%glimmix(data=test1
stmts=%str(
class classroom;
model III/NNN = D01 D02 D03 D04 D05 D06 D07

/ solution ddfm=bw;

random int D0&i/sub = classroom s;


ODS OUTPUT solutionr=blupz&i;
),
error = binomial,
link = logit
)
%end;
run;
%mend test;

Similar Threads

1. PROC GLIMMIX vs %GLIMMIX - Lost the convergence!

Hi everyone,

I used the %GLIMMIX macro before but have switched to the PROC GLIMMIX
recently to take advantage of the added features.

The problem is that for the same model and dataset, the convergence
criterion is met with the macro but not for the procedure...I get a
"DID NOT CONVERGE" message.  Anyone can tell what appears to be the
difference between my two programs?


Proc glimmix data=pairs4 ;
class ID_pair year;
model TARGETp (order=data)= REL1 DUMq MA_ACTIVITY SIZEp MBp LEVERAGEp
MGNTp EPp  / solution  dist=binomial link=logit error=binomial;
random intercept / subject=year;
run;


vs


%glimmix(data=pairs4,
   procopt=method=reml,

   stmts=%str(
      class ID_pair year;
      model TARGETp = REL1 DUMq MA_ACTIVITY SIZEp MBp LEVERAGEp MGNTp
EPp ;
random INTERCEPT/ SUBJECT=year; ),
  error=binomial,
   link=logit,
   solution
);
run;


What can explain the no convergence?

Thanks in advance,

Claudia

2. PROC GLIMMIX vs MACRO GLIMMIX

3. PROC GLIMMIX vs. MACRO GLIMMIX

4. %GLIMMIX vs. PROC GLIMMIX

5. Proc Glimmix without Random _residual_ VS Glimmix Macro

6. GLIMMIX vs. %GLIMMIX

7. Radial smoothing related question in Proc Glimmix

In longitudinal data, time trend is often curvilinear.  We usually model these nonlinear fixed time trends with polynomial function (quadratic, up to cubic), besides the other problems associated with higher order polynomials (multicollinearity, etc),  the results are hard to interpret for practical researchers.  So we are thinking about semiparametric mixed effects modeling approach, at least to model the FIXED time trend (average growth curve) nonparametrically in generalized mixed model framework.

New experimental SAS GLIMMIX procedure now is able to do some non-parametric things, such as univaraite and multivariate low-rank smoothing.

In example 6,  Radial Smoothing of Repeated Measures data, in  Proc Glimmix documentation,  they used radial smoothing (an approximate low-rank thin plate spline) to smooth individual time trend, but they treat this method as one type of covariance structure for random effect.

What I am interested in is  "mean response profile" or "average growth curve or fixed time effect."  So, my question is:  by smoothing individual time trend with radial smoothing covariance structure, do I also smoothed average (fixed) time trend?  I am aware of fixed time trend and individual random time trend are different things, but if I average individual predicted lines, this average line  is a smoothed line too even though we didn't actually smooth this average line (fixed time trend).

In other words, when you have nonlinear average growth curve, is it enough to just model

I don't know if I missed the boat entirely, but I appreciate your comments and help!

Thanks,

Haiyi

8. hierarchical models using proc glimmix question.