Allocation & Power Optimizer
Adaptive Bayes Two-Stage Drop the Losers Design
Authors: Alex Karanevich, Richard Meier, Stefan Graw
Department of Biostatistics, University of Kansas School of Medicine
View the
Instructions
Total sample size
Stage allocation optimization
Critical value α (one-sided)
Start point of total sample size
End point of total sample size
Step size
Effect size as
Best arm
Linear trend
Custom
Number of arms (including control)
Standardized effect size
Small (d=0.2)
Medium (d=0.5)
Large (d=0.8)
Custom
Standardized effect size
H
1
treatment means
(comma separated; control excluded)
Number of cores/threads
Advanced settings
Futility Stage 1: Stop trial early if
P(μ
Best
- μ
Ctrl
≥
δ
) ≤
ε
δ
= ?
ε
= ?
Number of n
1
and n
2
combinations per N
Post sample size
Simulation runs H
0
Simulation runs H
1
Random seed
Seed
Go!
Critical value α (one-sided)
Total sample size
Effect size as
Best arm
Linear trend
Custom
Number of arms (including control)
Standardized effect size
Small (d=0.2)
Medium (d=0.5)
Large (d=0.8)
Custom
Standardized effect size
H
1
treatment means
(comma separated; control excluded)
Number of cores/threads
Advanced settings
Futility Stage 1: Stop trial early if
P(μ
Best
- μ
Ctrl
≥
δ
) ≤
ε
δ
= ?
ε
= ?
Number of n
1
and n
2
combinations
Post sample size
Simulation runs H
0
Simulation runs H
1
Confidence level for simulated α and power
Simulation runs Dunnett/Bonferroni
Random seed
Seed
Go!
Show table