Help using Base SAS procedures

Error while using method=type3 in proc mixed

Accepted Solution Solved
Reply
Contributor
Posts: 26
Accepted Solution

Error while using method=type3 in proc mixed

Hi,

 

i got an error message while using method=type3 in proc mixed, saying that 'Error: Type 3 estimation is not available with this model.'.

i don't know what is the reason behind this error. well here i'm using Area under effective curve data and model was

 

proc mixed data=study.detectors7 method=type3;
classes sub trt;
model data= trt/ ddfm=sat;
random trt/type=FA0(2) sub=sub g;
repeated/group=trt sub=sub;
estimate 'T vs R' trt -1 1/cl alpha=0.1;
run;

 

version: SAS Enterprise guide 6.1


Accepted Solutions
Solution
‎09-19-2016 08:55 AM
SAS Super FREQ
Posts: 3,482

Re: Error while using method=type3 in proc mixed

Here is the documentation for PROC MIXED. Click on the "PROC MIXED" statement and scroll down to the METHOD= option. There you will see

The METHOD=TYPEn specifications apply only to variance component models with no SUBJECT= effects and no REPEATED statement. 

 

The error occurs because your model includes a REPEATED statement and subject effects.

 

If you want to model the repeated and random effects, consider using METHOD=REML (the default) or METHOD=ML  (maximum likelihood).

View solution in original post


All Replies
Solution
‎09-19-2016 08:55 AM
SAS Super FREQ
Posts: 3,482

Re: Error while using method=type3 in proc mixed

Here is the documentation for PROC MIXED. Click on the "PROC MIXED" statement and scroll down to the METHOD= option. There you will see

The METHOD=TYPEn specifications apply only to variance component models with no SUBJECT= effects and no REPEATED statement. 

 

The error occurs because your model includes a REPEATED statement and subject effects.

 

If you want to model the repeated and random effects, consider using METHOD=REML (the default) or METHOD=ML  (maximum likelihood).

☑ This topic is SOLVED.

Need further help from the community? Please ask a new question.

Discussion stats
  • 1 reply
  • 245 views
  • 2 likes
  • 2 in conversation