Bioequivalence and Bioavailability Forum

Sequence effect and 90% CI [General Sta­tis­tics]

posted by d_labes  – Berlin, Germany, 2009-12-10 10:24 (6032 d 19:48 ago) – Posting: # 4455
Views: 11,591

Dear Pankaj Mishra,

❝ ... Obviously if we remove the sequence effect, our total error variance will increase and consequently the intra-subject CV will rise which will further give us a wider 90% CI as compared to a model with "sequence" effect.

(emphasis by me)

This is definitely not true (at least not for a classical 2x2 cross-over).
Try an analysis (ANOVA) with effects formulation, period and subject.
Try a second analysis with effects formulation, period, sequence and subject within sequence.
Look at the result for the error term and the 90% confidence intervals of the formulation difference.
If all is ok with your software and model formulation you will get the same MSE and 90% CI from both models.

Try it!
And note that the sum of squares for sequence and subject within sequence of the second analysis sum up to the sum of squares of subject effects in the first.

Complete thread:

• Is the sequence effect mandatory? vezz 2009-12-09 21:52 [General Sta­tis­tics] 
  • Is the sequence effect mandatory? Pankaj Mishra 2009-12-10 07:21
    • Sequence effect and 90% CId_labes 2009-12-10 09:24
  • Is the sequence effect mandatory? d_labes 2009-12-10 09:01
    • Is the sequence effect mandatory? vezz 2009-12-10 11:02
      • Treatment by period interaction d_labes 2009-12-10 16:43
        • Old farts Helmut 2009-12-10 16:55
          • Old farts vezz 2009-12-10 18:10
            • New brooms Helmut 2009-12-10 19:19
            • Give sequence a chance d_labes 2009-12-11 09:42
              • Give history a chance Helmut 2009-12-11 13:40
          • Old farts? d_labes 2009-12-11 09:09
            • Young farts... Helmut 2009-12-11 13:51