Bioequivalence and Bioavailability Forum

Proc Mixed for >2 treatments [Software]

posted by H_Rotter – Germany, 2008-01-18 00:29 (6321 d 00:48 ago) – Posting: # 1522
Views: 12,508

Dear matz!

❝ I used both MIXED and GLM procedure to analyse the data of a 2x2 design with missing values. But both procedure gave me the same result (i.e; 90%C.I and ANOVA table).

As my subject line already said: 'Proc Miced for >2treatments'
In a 2x2 study there is no information which can be 'recovered' by using Proc MIXED - you simply have no test/reference ratio for drop-outs (incomplete samples). Therefore these subjects are omitted in both methods - giving identical results.

❝ So what is the advantage of using MIXED procedure for missing data set? ....Is there any particular reason for using MIXED procedure for analysing missing data sets?

Only for more than 2 treatments. Please reread my previous post and the linked ones as well...

Regards,
Hermann

Complete thread:

• Proc Mixed mathews 2008-01-08 07:47 
  • Proc Mixed for >2 treatments H_Rotter 2008-01-08 13:20
    • Proc Mixed for >2 treatments mathews 2008-01-17 08:02
      • Proc Mixed for >2 treatmentsH_Rotter 2008-01-17 23:29
        • Proc Mixed for >2 treatments mathews 2008-01-18 12:56
          • Proc Mixed for >2 Periods H_Rotter 2008-01-18 13:25