Precision of PowerTOST [Power / Sample Size]
Hi Helmut,
sorry if my thoughts were unclear in previous post.
As Astea has noted, 2 Laszlos in their paper estimated the required minimum sample size as 54 subjects for "2x3x3" design (CV=0.8, theta0=0.95). They claimed that "The precision of the estimation was evaluated by running the simulations twenty times".
My question was: is it possible for PowerTOST? Is it possible to get the mean power of 80% using the mean of 20 runs?
![[image]](img/uploaded/image191.jpg)
So the answer is: yes, that's possible, but you need to be unlucky (the probability is about 7%)
sorry if my thoughts were unclear in previous post.
As Astea has noted, 2 Laszlos in their paper estimated the required minimum sample size as 54 subjects for "2x3x3" design (CV=0.8, theta0=0.95). They claimed that "The precision of the estimation was evaluated by running the simulations twenty times".
My question was: is it possible for PowerTOST? Is it possible to get the mean power of 80% using the mean of 20 runs?
library("ggplot2")
library("PowerTOST")
reps <- 1E4
scABELoverall <- data.frame(rep = 1:reps)
scABEL20 <- data.frame(1:20)
for(external in 1:reps){
for(internal in 1:20){
scABEL20$power[internal] <- power.scABEL(CV=0.8, n=54, theta0=0.95, design="2x3x3", nsims=10000, setseed=F)
}
scABELoverall$meanpower[external] <- mean(scABEL20$power)
}
ggplot(scABELoverall, aes(meanpower))+
geom_density(fill = 2, alpha = 0.3)+
theme_bw()+
ggtitle(sprintf("%d reps: Mean is %.4f, SD is %.4f; %.2f %% of twenties are less than 0.8", reps, mean(scABELoverall$meanpower), sd(scABELoverall$meanpower), sum(scABELoverall$meanpower<0.8)/length(scABELoverall$meanpower)*100))
![[image]](img/uploaded/image191.jpg)
So the answer is: yes, that's possible, but you need to be unlucky (the probability is about 7%)

—
Kind regards,
Mittyri
Kind regards,
Mittyri
Complete thread:
- Sample Size Estimation -Replicate design- BE limits cakhatri 2017-11-30 09:44 [Power / Sample Size]
- Reference-scaling: Don’t use FARTSSIE! Helmut 2017-11-30 10:58
- Reference-scaling: Don’t use FARTSSIE? Astea 2017-11-30 22:50
- Reference-scaling: Don’t use FARTSSIE! Helmut 2017-12-01 11:42
- Reference-scaling: USE PowerTOST Astea 2017-12-01 14:16
- Reference-scaling: USE PowerTOST cakhatri 2017-12-04 06:00
- Reference-scaling: USE PowerTOST Helmut 2017-12-04 15:07
- Reference-scaling: USE PowerTOST cakhatri 2017-12-05 05:33
- Reference-scaling: USE PowerTOST Helmut 2017-12-04 15:07
- Use PowerTOST d_labes 2017-12-06 09:52
- Use PowerTOST, not Endrényi L, Tóthfalusi L tables mittyri 2017-12-07 16:03
- Precision in Endrényi L, Tóthfalusi L tables d_labes 2017-12-07 18:21
- Precision of PowerTOST mittyri 2017-12-08 22:56
- Precision of PowerTOST Helmut 2017-12-09 17:00
- Precision of PowerTOSTmittyri 2017-12-09 21:05
- Precision of PowerTOST Astea 2017-12-10 00:10
- Precision of PowerTOST Helmut 2017-12-10 20:27
- Precision of PowerTOST Astea 2017-12-10 21:13
- Precision of PowerTOST Helmut 2017-12-10 20:27
- Precision of PowerTOST Astea 2017-12-10 00:10
- Precision of PowerTOSTmittyri 2017-12-09 21:05
- Precision of PowerTOST Helmut 2017-12-09 17:00
- Precision of PowerTOST mittyri 2017-12-08 22:56
- Precision in Endrényi L, Tóthfalusi L tables d_labes 2017-12-07 18:21
- Use PowerTOST, not Endrényi L, Tóthfalusi L tables mittyri 2017-12-07 16:03
- Reference-scaling: USE PowerTOST cakhatri 2017-12-04 06:00
- Reference-scaling: USE PowerTOST Astea 2017-12-01 14:16
- Reference-scaling: Don’t use FARTSSIE! Helmut 2017-12-01 11:42
- Reference-scaling: Don’t use FARTSSIE! pjs 2018-02-09 13:08
- Reference-scaling: Don’t use FARTSSIE! Helmut 2018-02-09 20:15
- Reference-scaling: Don’t use FARTSSIE? Astea 2017-11-30 22:50
- Reference-scaling: Don’t use FARTSSIE! Helmut 2017-11-30 10:58