topic Re: Creating longitudinal dataset in SAS Programming

Creating longitudinal dataset

ChuksManuel — Mon, 13 Apr 2020 01:04:55 GMT

Hello programmers,

I am trying to implement this piece of code from a SAS guide. I can't implement it as it's giving me error messages.

Can anyone explain why it doesn't work?

 l=t(root(eps));      /* calculate cholesky root of cov matrix eps */  
 z=normal(j(1,n,seed));  /* generate n samplesize normals */    
xx=l*z*v;               /* multiply by cholesky root and variance */  
tx=t(xx);    
do i=1 to &n ; 
   Group[i] = round(ranuni(int(time()))); 
   x[i] = rannor(int(time()))*2 ;  
   do r=1 to &mm ;    
       b0[r]=-.2-2.6*cos((r-&m)*3.14159/ 30)* / );    
       yy[i,r]=b0[r]*Group[i]+3.5*sin(2*3.14159*x[i]/5);
     end;    
 end;   
 yy=yy+tx;               /*yy with multivariate normal error*/

Thanks

Re: Creating longitudinal dataset

RichardDeVen — Mon, 13 Apr 2020 01:30:32 GMT

And what are the errors ?

Re: Creating longitudinal dataset

ChuksManuel — Mon, 13 Apr 2020 02:58:37 GMT

Re: Creating longitudinal dataset

Kurt_Bremser — Mon, 13 Apr 2020 04:26:13 GMT

The comment mentions a matrix. Are you sure this is data step code, and not rather part of a proc iml?

Otherwise, it may be you are missing a custom function definition.

Re: Creating longitudinal dataset

Ksharp — Mon, 13 Apr 2020 11:00:04 GMT

It is IML code , not Data step .

You need SAS/IML module . Calling @Rick_SAS

Re: Creating longitudinal dataset

Rick_SAS — Mon, 13 Apr 2020 11:10:59 GMT

Start by using

PROC IML;

as the first line of the program. Then you need to define the covariance matrix, eps. You can either read it from a SAS data set into a matrix or you can input it manually:

proc iml;
eps = {
9 3   1,
3 4   1.3,
1 1.3 2};
 
...

Re: Creating longitudinal dataset

ChuksManuel — Mon, 13 Apr 2020 15:39:39 GMT

Thanks Dr Rick!