Help using Base SAS procedures

need Help with contrast statement

Accepted Solution Solved
Reply
Occasional Contributor
Posts: 9
Accepted Solution

need Help with contrast statement

I have a 3*6 (3 nitrogen rates and 6 rice varieties)  factorial design and I don't know how to make any contrast I need. Ive looked up SAS/STAT help but didn't find the answer. The questions are as follow:

comparison between Linear interaction of nitrogen with variety 6 vs. other varieties. ie. NL*(V6 vs. others). (NL means the linear interaction with Nitrogen factor).

NL*

NL*(V2 vs. V5)

NL*(V1 vs. V3 vs. V4)

Quadratic interactions are same as linear:

NQ*(V6 vs. others)

NQ*

NQ*(V2 vs. V5)

NQ*(V1 vs. V3 vs. V4)

The data are attached.

Thank you in advance

Attachment

Accepted Solutions
Solution
‎03-14-2014 08:51 AM
Respected Advisor
Posts: 2,655

Re: need Help with contrast statement

Posted in reply to EhsanKhedive

Some comments:

PROC GLM does not correctly calculate standard errors for split plot and strip plot designs.  Let me repeat that.  PROC GLM does not correctly calculate standard errors for split plot and strip plot designs.

Get a copy of SAS for Mixed Models, 2nd ed. by Littell et al., to see code for PROC MIXED for this design, and many examples of contrasts.

If you fit your data with a mixed model, consider using LSMESTIMATE statements (either in PROC GLIMMIX or PROC PLM), and it becomes straightforward to compare lsmeans of interest.

Steve Denham

View solution in original post


All Replies
Solution
‎03-14-2014 08:51 AM
Respected Advisor
Posts: 2,655

Re: need Help with contrast statement

Posted in reply to EhsanKhedive

Some comments:

PROC GLM does not correctly calculate standard errors for split plot and strip plot designs.  Let me repeat that.  PROC GLM does not correctly calculate standard errors for split plot and strip plot designs.

Get a copy of SAS for Mixed Models, 2nd ed. by Littell et al., to see code for PROC MIXED for this design, and many examples of contrasts.

If you fit your data with a mixed model, consider using LSMESTIMATE statements (either in PROC GLIMMIX or PROC PLM), and it becomes straightforward to compare lsmeans of interest.

Steve Denham

Occasional Contributor
Posts: 9

Re: need Help with contrast statement

Posted in reply to SteveDenham

Thank you very much, I found my way out with this book.

Best Regards

🔒 This topic is solved and locked.

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

Discussion stats
  • 2 replies
  • 185 views
  • 1 like
  • 2 in conversation