hirenpharm ☆ Ahmedabad, 2016-04-15 21:41 (2924 d 13:19 ago) Posting: # 16207 Views: 4,756 |
|
Hi Helmut, Can you please suggest when to use TOST (two one sided t-tests) for comparative bio availability in stead of a standard ANOVA approach using period, treatment, subjects as random effects? I was reading through https://sphhp.buffalo.edu/content/dam/sphhp/biostatistics/Documents/techreports/UB-Biostatistics-TR1107.pdf And also- two one-sided tests” (TOST) proposed by Schuirmann (1987) http://www.ncss.com/wp-content/themes/ncss/pdf/Procedures/PASS/Equivalence_Tests_for_Two_Means_using_Differences.pdf — Dr. Hiren Mehta |
Helmut ★★★ Vienna, Austria, 2016-04-18 16:02 (2921 d 18:58 ago) @ hirenpharm Posting: # 16213 Views: 3,654 |
|
Hi Hiren, ❝ Hi Helmut, ▲ Not interested in the opinions of other members? All guidelines I’m aware of recommend a 100(1–2α) confidence interval – not TOST. See also this post and the linked references. — 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 |
hirenpharm ☆ Ahmedabad, 2016-04-21 18:28 (2918 d 16:31 ago) @ Helmut Posting: # 16231 Views: 3,505 |
|
thanks Helmut. Obviously all members' opinion matters! |