topic Mixed models using restricted cubic splines in Statistical Procedures

Mixed models using restricted cubic splines

AGabriela — Mon, 03 Aug 2020 17:56:24 GMT

Hi there - The following code works but only with a very small sample size. My dataset is big and I get an ERROR message when I submit the code using the entire dataset. I have never used GLIMMIX, I'm not familiar with it. I wonder if this model can be run using MIXED instead? Would that help? Or are there any other ways to run this code more efficiently?

proc glimmix data=xx;

class studyid;

effect spl = spline(quarter / details naturalcubic basis=tpf(noint)knotmethod=percentiles(5));

model avgdose = spl /s ddfm=kr;

random int spl /subject = studyid type=vc solution; /*try vc instead of un*/

ods output parameterestimates=sf(keep=effect estimate

rename=(estimate=overall));

ods output solutionr=sr(keep=effect subject estimate

rename=(estimate=ssdev));

run;

Here is the error I get:

ERROR: Model is too large to be fit by PROC GLIMMIX in a

reasonable amount of time on this system. Consider

changing your model.

Re: Mixed models using restricted cubic splines

SteveDenham — Tue, 04 Aug 2020 11:02:42 GMT

Three questions: How much memory is available to you, how many quarters are you trying to fit and how many cases do you have (studyid's)? I suspect that your code is generating a lot of knots, and so it ends up with a too many parameters in the optimization compared to the available memory.

SteveDenham

Re: Mixed models using restricted cubic splines

Rick_SAS — Wed, 05 Aug 2020 12:48:15 GMT

Here's a suggestion: First fit a simple model that uses the main effects of QUARTER.

model avgdose = quarter /s ddfm=kr;

If the simple model converges and gives reasonable answers that you understand, then try the more complicated model that incorporates spline effects for QUARTER.

Re: Mixed models using restricted cubic splines

AGabriela — Wed, 05 Aug 2020 15:00:20 GMT

Thanks for your reply. That's always a good approach. I have done that, though.

Thanks again.

Re: Mixed models using restricted cubic splines

AGabriela — Wed, 05 Aug 2020 15:44:37 GMT

Thank you for your reply. Here are the answers:

1) On a regular basis just 32gb, but the IT department doubled that during the weekend. Unfortunately I didn't see a big improvement.

2) Nine quarters. I've tried 3, 4 or 5 knots. As expected, the smaller the number of knots the greater the chance the model will run.

3) If the dataset has 1k people (studyids) or less it runs fine. But unfortunately I have many more people (35k+) on the entire dataset.

Thanks again,

Gabriela