Randomized Allocation of Treatments in Sequential Experiments
Author:
- 1 July 1981
- journal article
- research article
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 43 (3) , 265-283
Abstract
Summary: Since the idea of sequential allocation was first studied, in a version of what is now called the multi-armed bandit problem, the results of many investigations have shown that, even when an optimal allocation rule can be found, its form depends on the parameters of the model and no single rule has emerged that is widely acceptable in practice. On the other hand, it can be shown that conventional methods, using equal allocations of the treatments under comparison, involve many unnecessary applications of inferior treatments even if the final choice of the “best” treatment relies on a sequential test. The evidence for this claim is based on an investigation of randomized allocation rules, where the choice of the next treatment is randomized in such a way that some preference is given to the “favourite” at any stage, but other treatments are not permanently excluded.Keywords
This publication has 19 references indexed in Scilit:
- Randomised allocation of treatments in sequential trialsAdvances in Applied Probability, 1980
- Bernoulli One-Armed Bandits--Arbitrary Discount SequencesThe Annals of Statistics, 1979
- The Two-Armed Bandit and the Controlled Clinical TrialJournal of the Royal Statistical Society: Series D (The Statistician), 1978
- Modified Two-Armed Bandit Strategies for Certain Clinical TrialsJournal of the American Statistical Association, 1978
- Finite Horizon Behavior of Policies for Two-Arm BanditsJournal of the American Statistical Association, 1974
- A two-stage procedure for choosing the better of two binomial populationsBiometrika, 1972
- Some Remarks on the Two-Armed BanditThe Annals of Mathematical Statistics, 1970
- Contributions to the "Two-Armed Bandit" ProblemThe Annals of Mathematical Statistics, 1962
- On Sequential Designs for Maximizing the Sum of $n$ ObservationsThe Annals of Mathematical Statistics, 1956
- Some aspects of the sequential design of experimentsBulletin of the American Mathematical Society, 1952