Help!
PROC GLIMMIX IS NOT CONVERGING! I get an error message
WARNING: Obtaining minimum variance quadratic unbiased estimates as starting values for the
covariance parameters failed. I first tried d=b link=clogit. My prof suggested deleting the d=b. My outcome is ordinal (1-4)
*****************ANALYSES*******************************;
PROC glimmix DATA=one order=data;
CLASS v6;
MODEL encourage= /SOLUTION link=clogit;
RANDOM intercept/SUBJECT=v6;
NLOPTIONS TECHNIQUE=NRRIDG;
weight v8;
TITLE 'STEP 1: RANDOM EFFECTS ANOVA';
run;
Well, the obvious first step would be to apply a multinomial distribution (dist=mult, or d=mult, with link=clogit).
This worked for me in translating the example in the PROC GENMOD documentation (Example 43.4 Ordinal Model for Multinomial Data) to a GLIMMIX framework.
Steve Denham
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.