topic Re: RANDOM EFFECT FAILS IN PROC MIXED in Statistical Procedures

RANDOM EFFECT FAILS IN PROC MIXED

Barkamih — Wed, 06 Mar 2019 15:26:12 GMT

HI all

I want to know why random effect didn't work in this proc mixed code,

When I run this code without the random effect it working perfectly

What do you think the main issue in this case?

this is what the log saying:

WARNING: ODS graphics with more than 5000 points have been suppressed. Use the PLOTS(MAXPOINTS= )
option in the PROC MIXED statement to change or override the cutoff.
NOTE: PROCEDURE MIXED used (Total process time):
real time 5.20 seconds
cpu time 5.12 seconds

1568! quit;

Regards

proc mixed data=have;
	class  pr mfd ca_month CAYEAR CA_EASE city ;
	model A =  pr mfd mfd*pr ca_month cayear ca_ease CITY;
	lsmeans  pr mfd  mfd*pr/pdiff adjust=tukey;
	random herd;
	lsmestimate MFD*pr 
	'MFD 1 v NMFD 1' 1 -1,
	'MFD 2 v NMFD 2' 0 0 1 -1,
	'MFD 3 v NMFD 3' 0 0 0 0 1 -1,
	'MFD 4 v NMFD 4' 0 0 0 0 0 0 1 -1 /adjust=bon;
run;quit;

Re: RANDOM EFFECT FAILS IN PROC MIXED

Rick_SAS — Wed, 06 Mar 2019 15:24:34 GMT

Please edit your question and add the error message in the SAS log.

Re: RANDOM EFFECT FAILS IN PROC MIXED

sld — Wed, 06 Mar 2019 21:03:17 GMT

What makes you think that the "random effect didn't work"?

You might try adding HERD to the CLASS statement. If the number of herds is large, then you can use HERD in the RANDOM statement as a continuous variable, but the dataset must be sorted appropriately by herd.

Re: RANDOM EFFECT FAILS IN PROC MIXED

Rick_SAS — Thu, 07 Mar 2019 00:42:27 GMT

That's the whole log? There are no errors. The WARNING just says that the ODS graphics will not automatically create scatter plots because the data contains more than 5000 points (the default).

You can modify the PROC MIXED statement to get the plots:

proc mixed data=have plots(maxpoints=none)=ALL;

...