Help using Base SAS procedures

Using PROC GLM to calculate an ANOVA for unbalanced data

Reply
New User
Posts: 1

Using PROC GLM to calculate an ANOVA for unbalanced data

Hello, I'm currently running a model very similar to this example:

 

https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_glm_sect053...

 

But I'm struggling to understand how the numbers 2.137, 2.2408 and 2.3162 are calculated in the Output 39.8.3 Expected Values of Type III Mean Squares. When using balanced data they should be all the same - the number of observations of each factor combination. But when we have unbalanced data the number of observations of each factor combination is different. Does anyone know how to calculate them or have any idea where they come from?

 

Thanks in advance!

Chloe

SAS Super FREQ
Posts: 3,477

Re: Using PROC GLM to calculate an ANOVA for unbalanced data

The mathematical formulas are in the section "Computation of Expected Mean Squares for Random Effects"

Ask a Question
Discussion stats
  • 1 reply
  • 92 views
  • 0 likes
  • 2 in conversation