
Title
Adaptive allocation and Anscombe's Theorem
Speaker
Vince Melfi, Michigan State University
Abstract
Adaptive allocation procedures, which allow the allocation of the current observation to depend on the allocations and responses of previous observations, have been developed in the context of clinical trials and industrial experiments. In this talk I will investigate the effect of the allocation scheme on the dependence structure of the data. The result of these investigations gives a simple method for proving consistency and asymptotic normality of estimators; the proof of asymptotic normality is based on Anscombe's Theorem.
Meet the speaker in Room 212 Cockins Hall at 4:30 p.m. Refreshments will be served.