Fig. 3
From: Bayesian dose selection design for a binary outcome using restricted response adaptive randomization
Under the alternative hypothesis of a linear treatment effect, the figure evaluates the effect of allocation between treatment and control, number of stages, and percentage of active subjects used in stage I on the global design properties: the probability of passing the Go, No-Go criteria and the probability of correct dose selection. Results are presented for four possible crude weights, I jk of the RAR specifications