Hodges Lehmann CI for a 2 Treat/3 Period Design [Nonparametrics]
Hi,
I have a 3 period, 2 treatment crossover (TRR, RTT) design and I've been asked to construct CI's for Tmax. Since the data is not normal, I plan to construct Hodges Lehmann CI's.
From reading Shein-Chung Chow, Jen-pei Liu. Design and Analysis of Bioavailability and Bioequivalence Studies Third Edition, it would appear I should calculate the pairwise calculations using the following formula:
Dik=1/4(2Yi1k - Yi2k - Yi3k) where i= number of differences & k= number of sequences.
Usually, Dik=Yi1 - Yi2
However, from looking at the SAS online help section, I cannot see a way of stipulating this using PROC NPAR1WAY. Do you have any suggestions?
Thanks!
Edit: Category changed. [Helmut]
I have a 3 period, 2 treatment crossover (TRR, RTT) design and I've been asked to construct CI's for Tmax. Since the data is not normal, I plan to construct Hodges Lehmann CI's.
From reading Shein-Chung Chow, Jen-pei Liu. Design and Analysis of Bioavailability and Bioequivalence Studies Third Edition, it would appear I should calculate the pairwise calculations using the following formula:
Dik=1/4(2Yi1k - Yi2k - Yi3k) where i= number of differences & k= number of sequences.
Usually, Dik=Yi1 - Yi2
However, from looking at the SAS online help section, I cannot see a way of stipulating this using PROC NPAR1WAY. Do you have any suggestions?
Thanks!
Edit: Category changed. [Helmut]
Complete thread:
- Hodges Lehmann CI for a 2 Treat/3 Period Designrandombadger 2011-05-31 20:26
- CI for a 2 Treat/3 Period Design Helmut 2011-05-31 21:06
- Hodges Lehmann CI for a 2 Treat/3 Period Design d_labes 2011-06-01 09:09
- Hodges Lehmann CI for a 2 Treat/3 Period Design randombadger 2011-06-01 19:32