Log in
Register|
Karthik ☆ Chennai, 2011-03-17 11:17 (5157 d 04:27 ago) Posting: # 6769 Views: 8,438 |
|
|
how to calculate power without using any software? or what are all the parameters we want consider to calculate power? Edit: Category and subject line changed. [Helmut] — regards, karthik |
|
Helmut ★★★   Vienna, Austria, 2011-03-28 16:40 (5146 d 00:05 ago) @ Karthik Posting: # 6824 Views: 7,473 |
|
|
Dear Karthik! ❝ how to calculate power without using any software? At least you need a pocket calculator and tables of the t-distribution.  For approximations see Hauschke D, Steinijans VW, Diletti E and M Burke Why don’t you use free software (FARTSSIE for Excel, package PowerTOST for R)? ❝ or what are all the parameters we want consider to calculate power? With given α (patient’s risk, commonly 5%), acceptance range (commonly 80–125%), expected T/R-ratio, and sample size you get power. If you want to calculate sample size for target power (e.g., 80%), increase the sample size until calculated power ≥ target power. Sample size calculation is always an iterative procedure. See one of my presentations. |
|
K2K ★ India, 2012-01-16 13:56 (4852 d 01:49 ago) @ Helmut Posting: # 7943 Views: 6,250 |
|
|
dear sir ❝ At least you need a pocket calculator and tables of the t-distribution. with the help of EXCEl sheet we can able to calculate power? if possible means advice me how to calculate. Sample Size Calculations …or the Myth of Power — thanks and regards Karthik |
|
Helmut ★★★ Vienna, Austria, 2012-01-16 14:14 (4852 d 01:30 ago) @ K2K Posting: # 7945 Views: 6,789 |
|
|
Dear Karthik! ❝ with the help of EXCEl sheet we can able to calculate power? Sure. Get David Dubins’ FARTSSIE.xls 1.7. You have to allow the execution of macros at startup. — 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 |
|
K2K ★ India, 2012-01-16 14:21 (4852 d 01:23 ago) @ Helmut Posting: # 7946 Views: 6,584 |
|
|
Dear sir! thanks without using that FARTSSIE.xls software with manual calculation from excel its possible to calculate power. http://bebac.at/lectures/Mumbai-BEWS2011-4.pdf in that pdf lecture page of 14 i try to calculate MSE but i didnt get the correct value sir but without multiply of 2 *(delcL/sqrt*tdist)^2 whether 2* (----) will come or not thanks you karthik — thanks and regards Karthik |
|
Helmut ★★★ Vienna, Austria, 2012-01-16 15:27 (4852 d 00:18 ago) @ K2K Posting: # 7949 Views: 6,595 |
|
|
Dear Karthik! Strange. In the following all intermediate steps (rounded values from the presentation and full internal precision; CLlo 0.91, CLhi 1.15, n1 11, n2 10): rounded full prec. (numerator)t 1.729 1.729131 (squared ×2!)CV 22.2% 22.1731%Quite close to R package PowerTOST: require(PowerTOST)FARTSSIE.xls ( Bioequivalence, Crossover) ISV from Published Point Estimate and CI:NA, 0.91, 1.15, 90, 21 ⇒ ISV = 22.2%— 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 |
|
K2K ★ India, 2012-01-18 10:35 (4850 d 05:10 ago) @ Helmut Posting: # 7958 Views: 6,286 |
|
|
dear sir i calculated N=28 (sqrt(1/28+1/28)) so only i didnt get sir sorry and thanks for your vauable reply — thanks and regards Karthik |
Ing. Helmut Schütz