Fig. 1From: Adaptive treatment allocation and selection in multi-arm clinical trials: a Bayesian perspectiveEffect of the choice of the design parameters ε and δ in BARTA on the number of patients allocated to the experimental treatment and on the total number of treatment successes. Cumulative distribution functions of N1(200) (top) and S(200) (bottom) are shown, based on 5000 simulated data sets, under \(\mathbb {Q}_{null}\) with true parameter values θ0=θ1=0.3 and \(\mathbb {Q}_{alt}\) with values θ0=0.3,θ1=0.5. Three combinations of the design parameters were used: (a) ε=0.1,δ=0.1, (b) ε=0.05,δ=0.1, (c) ε=0.2,δ=0.05. In addition, (d) represents a completely symmetric treatment allocation. For comparison we also plot the corresponding CDF under the alternative hypothesis obtained by using fractional Thompson’s rule with respective parameters κ=0.25,0.5,0.75 and 1Back to article page