On the asymptotic power of the statistical test under Response-Adaptive randomization

We analyze the impact of Response-Adaptive (RA) randomization rules in the case of normally response trials when the aim consists in testing the superiority of one of two available treatments. Taking into account the asymptotic power of the classical statistical test, we show that some target allocations may induce consistent loss of power. Moreover we analyze the conditions under which this problem can be avoided, adopting suitable targets as well as correct applications of RA rules.