BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
tompatmck
Fluorite | Level 6

Hello, 

 

I'm analyzing data collected from a Randomized Complete Block Design with missing observations, so I'm using Proc mixed (SAS 9.4). Within each block there is one fixed main plot factor (A) and one fixed subplot factor within each plot (B). I have been analyzing as a split-plot design with block as a random effect:

 

proc mixed;

class block A B;

model Y = A B A*B;

random block block*A;

 

Is this correct?

 

Also, I have another response where I took repeated measurements over time (time).  It is also unbalanced.  I have been using:

 

proc mixed;

class time block A B;

model Y = time|A|B;

random block block*A;

repeated time/ subject= block*A*B type=un ddfm=KR;

 

I've also been trying different covariance models with the repeated measures.  With the Unstructured and Compound Symmetry covariance structure, the random statement shouldn't include the subject effect.  Is the above random statement OK because they are different?

 

Thanks.

1 ACCEPTED SOLUTION

Accepted Solutions
sld
Rhodochrosite | Level 12 sld
Rhodochrosite | Level 12

@tompatmck wrote:

Hello, 

 

I'm analyzing data collected from a Randomized Complete Block Design with missing observations, so I'm using Proc mixed (SAS 9.4). Within each block there is one fixed main plot factor (A) and one fixed subplot factor within each plot (B). I have been analyzing as a split-plot design with block as a random effect:

 

proc mixed;

class block A B;

model Y = A B A*B;

random block block*A;

 

Is this correct? 

 

Yes, this model seems to be consistent with your description of the experimental design.

 


Also, I have another response where I took repeated measurements over time (time).  It is also unbalanced.  I have been using:

 

proc mixed;

class time block A B;

model Y = time|A|B;

random block block*A;

repeated time/ subject= block*A*B type=un ddfm=KR;

 

I've also been trying different covariance models with the repeated measures.  With the Unstructured and Compound Symmetry covariance structure, the random statement shouldn't include the subject effect.  Is the above random statement OK because they are different?

 

 

The above code is correct except that "ddfm=KR" is an option on the MODEL statement, not the REPEATED statement. I don't know what you mean by "because they are different"; you could provide more information about your question.

 

I hope this helps.

 

View solution in original post

1 REPLY 1
sld
Rhodochrosite | Level 12 sld
Rhodochrosite | Level 12

@tompatmck wrote:

Hello, 

 

I'm analyzing data collected from a Randomized Complete Block Design with missing observations, so I'm using Proc mixed (SAS 9.4). Within each block there is one fixed main plot factor (A) and one fixed subplot factor within each plot (B). I have been analyzing as a split-plot design with block as a random effect:

 

proc mixed;

class block A B;

model Y = A B A*B;

random block block*A;

 

Is this correct? 

 

Yes, this model seems to be consistent with your description of the experimental design.

 


Also, I have another response where I took repeated measurements over time (time).  It is also unbalanced.  I have been using:

 

proc mixed;

class time block A B;

model Y = time|A|B;

random block block*A;

repeated time/ subject= block*A*B type=un ddfm=KR;

 

I've also been trying different covariance models with the repeated measures.  With the Unstructured and Compound Symmetry covariance structure, the random statement shouldn't include the subject effect.  Is the above random statement OK because they are different?

 

 

The above code is correct except that "ddfm=KR" is an option on the MODEL statement, not the REPEATED statement. I don't know what you mean by "because they are different"; you could provide more information about your question.

 

I hope this helps.

 

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 2316 views
  • 0 likes
  • 2 in conversation