Log in
RegisterPotvin C consequences [Two-Stage / GS Designs]
posted by Helmut 
– Vienna, Austria, 2022-06-27 16:49 (661 d 12:30 ago) – Posting: # 23094
Views: 1,973
Hi Mittyri,
Yes, though not only on the CV but on n1 as well (though less so).
In principle yes but I received a few deficiency letters asking for a “post hoc assessement of the Type I Error”. That’s not well-thought-out. Of course, we know n1 but the CV is just an estimate. The true one (used in the simulations of the framework) is unknown.
There are no problems with “Method B”. In the entire framework of CV/n1-combinations the T1E is controlled. When you explore a narrow grid (step size of CV 2% and of n1 2) you will get this (noncentral t-distribution instead of the shifted central t like in the paper):
As you already discovered, “Method C” can be problematic – though only in certain rare cases.
IMHO, it was unfortunate by Potvin et al. to select the same adjusted α in both methods. Any ‘Type 2’ TSD will require more adjustment than its ‘Type 1’ counterpart.
BTW, 0.0294 in “Method B” is overly conservative; 0.0302 controls the Type I Error as well. On the other hand, for “Method C” you need 0.0282.
Consider to avoid this stuff in the future. See this post and followings for better alternatives.
❝ OK, what about this:
❝ power.tsd(method = "C", CV = 0.2, n1 = 12, theta0 = 1.25)
❝ p(BE) = 0.0511
❝ power.tsd(method = "C", CV = 0.3, n1 = 12, theta0 = 1.25)
❝ p(BE) = 0.043743
❝
❝ So T1E (OK, maximum type I error) depends on the CV, isn't it?
Yes, though not only on the CV but on n1 as well (though less so).
❝ Or may be I misunderstood your point and we should not verify T1E during estimation process since we are assuming that the framework was stated as acceptable in the Protocol?
In principle yes but I received a few deficiency letters asking for a “post hoc assessement of the Type I Error”. That’s not well-thought-out. Of course, we know n1 but the CV is just an estimate. The true one (used in the simulations of the framework) is unknown.
There are no problems with “Method B”. In the entire framework of CV/n1-combinations the T1E is controlled. When you explore a narrow grid (step size of CV 2% and of n1 2) you will get this (noncentral t-distribution instead of the shifted central t like in the paper):
![[image]](img/uploaded/image832.png)
Maximum Type I Error 0.048856 at CV 24% and n1 12.
As you already discovered, “Method C” can be problematic – though only in certain rare cases.
![[image]](img/uploaded/image833.png)
Maximum Type I Error 0.051250 at CV 22% and n1 12
(the yellow dot). Thick red contour at the bottom encloses T1E > 0.05.
IMHO, it was unfortunate by Potvin et al. to select the same adjusted α in both methods. Any ‘Type 2’ TSD will require more adjustment than its ‘Type 1’ counterpart.
BTW, 0.0294 in “Method B” is overly conservative; 0.0302 controls the Type I Error as well. On the other hand, for “Method C” you need 0.0282.
Consider to avoid this stuff in the future. See this post and followings for better alternatives.
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png)
Helmut Schütz
![[image]](https://static.bebac.at/img/CC by.png)
The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz
The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Complete thread:
- T1E after 1st stage in Power2Stage mittyri 2022-06-24 13:43 [Two-Stage / GS Designs]
![[Mix]](https://static.bebac.at/img/mix.png "open in Mix view")
- T1E after 1st stage in TSDs Helmut 2022-06-24 14:33
- T1E after 1st stage in TSDs mittyri 2022-06-24 21:12
- T1E after 1st stage in TSDs Helmut 2022-06-25 14:07
- Potvin C consequences mittyri 2022-06-25 23:29
- Potvin C consequences ElMaestro 2022-06-26 09:11
- Potvin C consequences mittyri 2022-06-27 13:45
- Potvin C consequencesHelmut 2022-06-27 14:49
- from Potvin C to Inverse Normal mittyri 2022-06-27 17:13
- From Potvin C to Inverse Normal: history (and outlook?) Helmut 2022-06-27 18:54
- from Potvin C to Inverse Normal mittyri 2022-06-27 17:13
- Potvin C consequences ElMaestro 2022-06-27 17:13
- true CV ≠ observed CV mittyri 2022-06-27 17:22
- Potvin C consequencesHelmut 2022-06-27 14:49
- Potvin C consequences mittyri 2022-06-27 13:45
- Potvin C consequences ElMaestro 2022-06-26 09:11
- Potvin C consequences mittyri 2022-06-25 23:29
- T1E after 1st stage in TSDs Helmut 2022-06-25 14:07
- T1E after 1st stage in TSDs mittyri 2022-06-24 21:12
- T1E after 1st stage in Power2Stage ElMaestro 2022-06-24 18:25
- T1E after 1st stage in Power2Stage mittyri 2022-06-24 21:05
- T1E after 1st stage in Power2Stage ElMaestro 2022-06-24 22:55
- T1E after 1st stage in Power2Stage mittyri 2022-06-24 21:05
- T1E after 1st stage in TSDs Helmut 2022-06-24 14:33
Ing. Helmut Schütz