Log in
RegisterDefining alpha for replicate design [Regulatives / Guidelines]
Dear All,
As per the code defined in US FDA guideline we define ALPHA as 0.10 in ESTIMATE statement of PROC MIXED for replicate study designs.
While we do not define any such statement and use default ALPHA = 0.05 for non-replicate crossover studies with PROC GLM or MIXED procedures.
In both type of studies (replicate or non-replicate), we need to present 90% CI interval.
Can anybody please tell me the rationale behind this?
Thanks in advance...
KR
As per the code defined in US FDA guideline we define ALPHA as 0.10 in ESTIMATE statement of PROC MIXED for replicate study designs.
While we do not define any such statement and use default ALPHA = 0.05 for non-replicate crossover studies with PROC GLM or MIXED procedures.
In both type of studies (replicate or non-replicate), we need to present 90% CI interval.
Can anybody please tell me the rationale behind this?
Thanks in advance...
KR
Complete thread:
- Defining alpha for replicate designKR 2009-09-15 14:34
![[Mix]](https://static.bebac.at/img/mix.png "open in Mix view")
- Defining alpha for replicate design sinhmar 2009-09-19 10:43
Ing. Helmut Schütz