I was running a Proc Mixed model a test using code from the SAS website. I found that the solution option for the random statement produced the random effects coefficients (intercept and slope) but also produced a standard error of the prediction that was then used to calculate a t statistic for both coefficients. I have never seen this anywhere else and it seems that calculating the t stat might be useful. Can anyone say exactly how the standard error of the predictions was calculated? The manual is very vague.
Since you are interested in Mixed Models, you need to get the book "SAS for Mixed Models, 2nd edition" by Littelll et al. (2006). Lots on random effects. There will be a third edition coming out in 2017 or 2018.
This is actually standard practice. Except for some very simple situations, there is no straight-forward scalar equation for the standard error for the predicted random effect coefficient (really should be called square root of the variance of the prediciton). You typically can only write this using matrix notation. In the User's Guide, this is described under Estimating Fixed and Random Effects in the Mixed Model in the Theory Section (see the C^ matrix). The relevant section is for gamma^ - gamma vector.
Since you are interested in Mixed Models, you need to get the book "SAS for Mixed Models, 2nd edition" by Littelll et al. (2006). Lots on random effects. There will be a third edition coming out in 2017 or 2018.
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
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.