Power parallel group: PowerTOST/SAS/PASS [🇷 for BE/BA]
Dear Helmut,
As you might imagine: that makes me (as a nitpicker) a little bit nervous.
But of course I trust in my software even if I myself had written it. I only don't trust in any computers
.
Fired up my beasty beast
.
Used the prudish beauty
Let's look at the results (5 decimals, PASS 14 from your link):
Note especially the differences between help page and actually calculated for PASS12.
Victory all a long the line
! PowerTOST must be correct because SAS says so
.
"Enjoy the best sample size software on the market" (PASS12 startup screen).
Power: That which statisticians are always calculating but never have.
Guernsey McPerson's "The Devil's Drug Development Dictionaries"
❝ Occasionally slight differences for the parallel design example.
As you might imagine: that makes me (as a nitpicker) a little bit nervous.
But of course I trust in my software even if I myself had written it. I only don't trust in any computers

Fired up my beasty beast
![[image]](img/uploaded/image1.gif)
Used the prudish beauty

Proc Power;
twosamplemeans
test=equiv_ratio
power=. /* means we solve for power */
alpha=0.05
meanratio=1 1.05
npergroup=50 150 250 350 450 550
cv=1.5
lower=0.8
upper=1.25
;
run;
Let's look at the results (5 decimals, PASS 14 from your link):
------ power ------
N1 N2 N R1 COV RL RU Alpha PowerTOST SAS PASS14 PASS12(help) PASS12(calc)
50 50 100 1.00 1.5 0.8 1.25 0.05 0.00000 0.00000 0.00000 0.0000 0.00000
150 150 300 1.00 1.5 0.8 1.25 0.05 0.10488 0.10488 0.10488 0.1049 0.10488
250 250 500 1.00 1.5 0.8 1.25 0.05 0.48431 0.48431 0.48431 0.4843 0.48431
350 350 700 1.00 1.5 0.8 1.25 0.05 0.71606 0.71606 0.71704 0.7170 0.71704
450 450 900 1.00 1.5 0.8 1.25 0.05 0.84896 0.84896 0.84945 0.8494 0.84945
550 550 1100 1.00 1.5 0.8 1.25 0.05 0.92185 0.92185 0.92208 0.9221 0.92208
50 50 100 1.05 1.5 0.8 1.25 0.05 0.00000 0.00000 0.00000 0.0000 0.00000
150 150 300 1.05 1.5 0.8 1.25 0.05 0.09731 0.09731 0.09731 0.1010 (!) 0.09731
250 250 500 1.05 1.5 0.8 1.25 0.05 0.43421 0.43421 0.43421 0.4360 (!) 0.43421
350 350 700 1.05 1.5 0.8 1.25 0.05 0.63561 0.63561 0.63661 0.6366 0.63661
450 450 900 1.05 1.5 0.8 1.25 0.05 0.75960 0.75960 0.76018 0.7602 0.76018
550 550 1100 1.05 1.5 0.8 1.25 0.05 0.83925 0.83925 0.83961 0.8396 0.83961
Note especially the differences between help page and actually calculated for PASS12.
Victory all a long the line


"Enjoy the best sample size software on the market" (PASS12 startup screen).
Power: That which statisticians are always calculating but never have.
Guernsey McPerson's "The Devil's Drug Development Dictionaries"
—
Regards,
Detlew
Regards,
Detlew
Complete thread:
- Power calculation BE-proff 2015-08-13 13:44 [🇷 for BE/BA]
- Power calculation Helmut 2015-08-13 13:59
- Power calculation BE-proff 2015-08-13 14:44
- Example(s) Helmut 2015-08-13 14:58
- Power parallel group: PowerTOST/SAS/PASSd_labes 2015-08-14 10:29
- Power parallel group: PowerTOST/SAS/PASS Helmut 2015-08-14 15:34
- Power parallel group: PowerTOST/SAS/PASSd_labes 2015-08-14 10:29
- Slippery ground d_labes 2015-08-14 10:50
- Advanced example Helmut 2015-08-14 16:41
- Advanced example for 2X2X2 mittyri 2015-08-14 19:40
- Spaghetti viennese Helmut 2015-08-15 03:22
- Spaghetti viennese d_labes 2015-08-17 08:45
- Spaghetti viennese Helmut 2015-08-15 03:22
- Advanced example BE-proff 2015-08-17 08:55
- Advanced example for 2X2X2 mittyri 2015-08-14 19:40
- Advanced example Helmut 2015-08-14 16:41
- Example(s) Helmut 2015-08-13 14:58
- Power calculation BE-proff 2015-08-13 14:44
- Power calculation Helmut 2015-08-13 13:59