Flexible Algorithms for Creating and Analyzing Adaptive Sampling Procedures

We describe a collection of algorithms and techniques that have been developed to aid in the design and analysis of adaptive allocation procedures. The emphasis is on providing flexibility to the investigator, so that appropriate statistical and practical concerns can be addressed directly. The techniques described allow for optimizations previously not attainable. They also permit exact evaluations for a wide range of criteria and are intended to encourage investigators to explore more alternatives. Optimizations investigated include 2and 3-population fully sequential models, few-stage models, and models with constrained switching between options. One of our algorithmic approaches, path induction, speeds up the process of evaluating a procedure multiple times so that thorough robustness studies can be undertaken. Our approaches can be utilized with both Bayesian and frequentist analyses.

[1]  P F Thall,et al.  A two-stage design for choosing among several experimental treatments and a control in clinical trials. , 1989, Biometrics.

[2]  Shen Zheng,et al.  ASYMPTOTICALLY RISK EFFICIENT TWO STAGE PROCEDURE FOR ESTIMATING THE PRODUCT OF k(>= 2) MEANS , 1998 .

[3]  D. Coad.  SEQUENTIAL ALLOCATION INVOLVING SEVERAL TREATMENTS , 1995 .

[4]  Quentin F. Stout,et al.  Optimal Adaptive Equal Allocation Rules , 1992 .

[5]  D. Berry,et al.  Adaptive assignment versus balanced randomization in clinical trials: a decision analysis. , 1995, Statistics in medicine.

[6]  R. Simon,et al.  Adaptive treatment assignment methods and clinical trials. , 1977, Biometrics.

[7]  Quentin F. Stout,et al.  Optimal allocation for estimating the mean of a bivariate polynomial , 1996 .

[8]  Donald A. Berry,et al.  Bandit Problems: Sequential Allocation of Experiments. , 1986 .

[9]  P. Armitage.  The search for optimality in clinical trials , 1985 .

[10]  Peter Hall,et al.  Asymptotic Theory of Triple Sampling for Sequential Estimation of a Mean , 1981 .

[11]  Quentin F. Stout,et al.  Determining optimal few stage allocation rules , 1995 .

[12]  Anne S. Hawkins.  Bandit Problems—Sequential Allocation of Experiments , 1987 .

[13]  D. Teneketzis,et al.  Asymptotically efficient adaptive allocation rules for the multiarmed bandit problem with switching cost , 1988 .

[14]  R. N. Bradt.  On the Design and Comparison of Certain Dichotomous Experiments , 1954 .

[15]  You-Gan Wang.  Sequential allocation in clinical trials , 1991 .

[16]  Quentin F. Stout,et al.  Adaptive Allocation in the Presence of Missing Outcomes , 1998 .

[17]  Q. Stout,et al.  Bandit Strategies for Ethical Sequential Allocation , 1992 .

[18]  Quentin F. Stout,et al.  Optimal Few-Stage Allocation Procedures , 1998 .

[19]  Quentin F. Stout,et al.  Exact computational analyses for adaptive designs , 1995 .

[20]  Quentin F. Stout,et al.  A Parallel Program for 3-Arm Bandits , 1997 .

[21]  Quentin F. Stout,et al.  Sequential Allocation with Minimal Switching , 1996 .

[22]  H. Büringer,et al.  Nonparametric Sequential Selection Procedures , 1980 .

[23]  Quentin F. Stout,et al.  Path induction for evaluating sequential allocation procedures , 1999 .

[24]  Rebecca A. Betensky,et al.  An O'Brien-Fleming sequential trial for comparing three treatments , 1996 .

[25]  Micha Hofri,et al.  On the Optimal Control of Two Queues with Server Setup Times and its Analysis , 1987, SIAM J. Comput..

[26]  C. P. Shapiro,et al.  Allocation Schemes for Estimating the Product of Positive Parameters , 1985 .

[27]  R. Bellman.  A PROBLEM IN THE SEQUENTIAL DESIGN OF EXPERIMENTS , 1954 .

[28]  N. Schmitz.  Optimal Sequentially Planned Decision Procedures , 1992 .