Bioequivalence and Bioavailability Forum

Interesting! [General Statistics]

posted by ElMaestro – Denmark, 2017-03-28 23:10 (3378 d 01:18 ago) – Posting: # 17195
Views: 12,231

Hi d_labes,

❝ p-value of treatment effect = 0.076998 (not significant at the 5% level)

❝ 90% CI = 90.76 ... 99.62% (doesn't contain 100%)

Thank you.
This somehow does rock my fundamental understanding.
I need to sit down with pen and paper and work it out.

If anova p<0.05 for treatment then the 90% CI does not include 1.00. But vice versa isn't the case?!??? This is not at all intuitive to me since the Null is equality at alpha 5%. I must look a bit into this, I believe.

—
Pass or fail!
ElMaestro

Complete thread:

Relationship between calculated 90% CI and sign. treatment effect in BE GM 2017-03-27 19:40 [General Statistics]
- 1–2α CI and TOST at α 0.05 Helmut 2017-03-27 23:47
  - 1–2α CI and TOST at α 0.05 GM 2017-03-28 06:50
    - 1–2α CI and TOST at α 0.05 DavidManteigas 2017-03-28 11:48
      - 1–2α CI and TOST at α 0.05 ElMaestro 2017-03-28 13:12
        
        alpha TOST is not alpha 2-sided d_labes 2017-03-28 15:12
        
        Interesting!ElMaestro 2017-03-28 21:10
        
        Google has the answer d_labes 2017-03-29 08:20
        Interesting! DavidManteigas 2017-03-29 11:16
        
        Interesting! ElMaestro 2017-03-29 11:20
        
        Interesting! DavidManteigas 2017-03-29 12:28
        
        95% CI for a test on difference d_labes 2017-03-29 14:24
        Interesting! GM 2017-03-29 20:06
        
        Interesting! nobody 2017-03-30 08:24
      - Relationship between calculated 90% CI and sign. treatment effect in BE GM 2017-03-29 12:11
        
        Relationship between calculated 90% CI and sign. treatment effect in BE DavidManteigas 2017-03-29 12:30

Interesting! [General Sta­tis­tics]

Complete thread:

Interesting! [General Statistics]