P. Brutti , F. De Santis and S. Gubbiotti
A standard Bayesian stopping rule for sequential trials is based on the posterior probability that a treatment e ect exceeds a minimum relevant clinical threshold. In this paper we consider a robust version of this criterion by replacing the single prior distribution with a class of prior distributions. We compare the average sample sizes of the robust sequential approach both with the sample sizes of the non robust approach and of the non sequential approach. A surprising result is that, in some cases, the average sample sizes of the robust sequential approach are smaller than the non sequential sample sizes.
Clinical trials, contamination priors, Robustness, Sample size determination, Sequential analysis
Tipo di pubblicazione: