Bibliography

      Stopping rules for proofreading, (with T. Ferguson) (1989). Journal of Applied Probability 26: 304-313.

      Recent progress in clinical trial designs that adapt for ethical purposes, (1989). (Discussion) Statistical Science 4: 327-336.

      Computational problems associated with minimizing the risk in a simple clinical trial, (1989). Contemporary Mathematics: Statistical Multiple Integration. American Mathematics Association 115: 239-257. (N. Flournoy and R. Tsutakawa, ed.'s).

      Sequential allocation for an estimation problem with ethical cost, (with M. Woodroofe). (1990) Annals of Statistics 18: 1358-1367.

      Sequential selection of a relatively best object, (with T. Ferguson and M. Tamaki) (1991). Contemporary Mathematics: Sequential Search and Selection in Real Time, American Mathematics Association 125 37-58. (T. Ferguson and S. Samuels, ed.'s).

      Report of the New Researchers' Committee, (with NRC) (1991). Statistical Science, 6: 163-174.

      Rejoinder to Readers' Comments to the New Researchers' Committee Report, (with NRC) (1992). Statistical Science, 7: 265-266.

      Using historical controls in clinical trials: Application to ECMO (with Donald Berry), (1994). Fifth Purdue International Symposium on Statistical Decision Theory and Related Topics. V; 141-156.

      Exact computational analyses for sequential allocation problems, (with Q. F. Stout) (1995). Adaptive Designs: Institute of Mathematical Statistics Lecture Notes 25 65-87. (N. Flournoy and W. F. Rosenberger, ed.'s).
      Keywords: constrained dynamic programming, forward induction, backward induction, Bayesian design, multiple criteria, clinical trials. [Abstract | Paper (Postscript) | Paper (PDF)]

      A modified bandit as an approach to ethical allocation in clinical trials, (1995). Adaptive Designs: Institute of Mathematical Statistics Lecture Notes 25 223-237. (N. Flournoy and W. F. Rosenberger, ed.'s). [Abstract]

      Asymptotic expansions for randomly censored survival data, (1996). Statistics and Probability Letters 26 187-197.
      Keywords: parameter transformations, posterior distributions, normal likelihood. [Abstract]

      Optimal allocation for estimating the mean of a bivariate polynomial, (with Q. F. Stout) (1996). Sequential Analysis 15, pp. 71-90.
      Keywords: nonlinear estimation, sequential design, robustness, myopic, hyperopic, dynamic programming, adaptive allocation, product, Bayesian, fault tolerance. [Abstract | Paper with color figures | Paper with grayscale figures]

      Sequentially deciding between two experiments for estimating a common success probability (with C. Page and Q. F. Stout) (1998). J. American Statistical Association, 93, pp. 1502-1511.
      Keywords: batch testing, risk assessment, infection rate, grouped data, omniscient allocation, composite testing. [Abstract | Paper (Postscript) | Paper (PDF)]

      Flexible algorithms for creating and analyzing adaptive designs (with Q. F. Stout), New Developments and Applications in Experimental Design, Institute of Math. Stat. Lecture Notes -- Monograph Series, 34 (1998), pp. 91-105.
      Keywords: sequential allocation, dynamic programming, forward induction, bandit problems, algorithms, high-performance computing, stochastic optimization, parallel computing. [Abstract | Paper (Postscript) | Paper (PDF)]

      Using path induction to evaluate sequential allocation procedures (with Q. F. Stout), SIAM J. Scientific Computing 21 (1999), pp. 67-87.
      Keywords: backward induction, adaptive allocation, forward induction, staged allocation, group sampling, path counting, Bayesian, bandit problems, dynamic programming, stochastic optimization. [Abstract | Paper (Postscript) | Paper (PDF)]

      A program for sequential allocation of three Bernoulli populations (with R. Oehmke and Q. F. Stout), Computational Statistics and Data Analysis 31 (1999), pp. 397-416.
      Keywords: adaptive sampling, bandit, estimation, dynamic programming, path induction, parallel computing, design of experiments, high performance computing. [Abstract | Paper (Postscript) | Paper (PDF)]

      Optimizing a unimodal response function for binary variables (with Q. F. Stout), Optimum Design 2000, A. Atkinson, B. Bogacka, and A. Zhigljavsky, eds., Kluwer, 2001, pp. 195-208.
      Keywords: nonparametric, shape-constrained, unimodal regression, bandit, approximate Gittins index, stochastic approximation, Polya urn, up and down, Markov chain, random walk. [Abstract | Paper (Postscript) | Paper (PDF)]

      Optimal adaptive designs for delayed response models: exponential case (with R. Oehmke and Q.F. Stout), MODA6: Model Oriented Data Analysis, A. Atkinson, P. Hackl, W. Mueller, eds., Physica Verlag, 2001, pp. 127-134.
      Keywords: delayed response, sequential analysis, dynamic programming, computational learning theory, multi-arm bandit models, parallel computing, clinical trial, recursive equations, design of experiments, medical ethics, clinical trial. [Abstract | Paper (Postscript) | Paper (PDF)]

      Scalable algorithms for adaptive statistical designs (with R. Oehmke and Q.F. Stout), Proc. SC 2000 (Supercomputing), 16 p. Reprinted in Scientific Programming 8 (2001), pp. 183-193.
      Keywords: dynamic programming, computational learning theory, multi-arm bandit models, message-passing, dynamic domain decomposition, memory-intensive computing, load balancing, sequential analysis, performance analysis, experimental algorithms, parallel computing, clinical trial, recursive equations, design of experiments, medical ethics, clinical trial, delayed response, backward induction, MPI, IBM SP, OpenMP, SGI Origin. [Abstract | Paper.ps | Paper.pdf]

      Optimal few-stage designs (with Q.F. Stout). Journal Statistical Planning and Inference 104 (2002), pp. 121-145.
      Keywords: two-stage, three-stage, clinical trial, medical ethics, group allocation, selection, estimation, bandit models, product of means. [Abstract | Paper.ps | Paper.pdf]

      Directed walk designs for dose response problems with competing failures modes (with M.C. Meyer and Q.F. Stout) (2003). Biometrics, 59 pp. 229-236.
      Keywords: adaptive sampling, phase I/II clinical trial, sequential allocation, random walk, shape-constrained regression, parametric, nonparamentric, Bayesian, frequentists, smooothed, curve estimation, design of experiments, toxic. [Abstract | Paper.ps | Paper.pdf]

      Optimal screening designs with flexible cost and constraint structures (with Q.F. Stout) (2005). Journal of Statistical Planning and Inference 132, pp. 149-162.
      Keywords: drug testing, acceptance-rejection sampling, staged procedure, stopping rule, dynamic programming, design of experiments, decision theory, backward induction. [Abstract | Paper.pdf]

      Parallel programs for adaptive designs (with Q.F. Stout) (2006). Handbook of Parallel Computing and Statistics, E. Kontoghiorghes, ed., Marcel Dekker, pp. 347-373.
      Keywords: multi-arm bandit models, dynamic programming, speedup, clinical trial, delayed response, response adaptive sampling, parallelizing recursive equations, sequential analysis. [Abstract | Paper.pdf]

      New adaptive designs for delayed response models (with R. Ohemke and Q.F. Stout) (2006). Journal of Statistical Planning and Inference, 136, pp. 940-1955.
      Keywords: randomized play-the-winner, hyperopic design, bandit problem. [Abstract | Paper.pdf]

      Response adaptive designs that incorporate switching costs and constraints (with Q.F. Stout) (2007). Journal of Statistical Planning and Inference, 137, pp. 2654-2665
      Keywords: switching costs, constraints, hyperopic procedure, bandit problem, nonlinear Bayes risk, dynamic programming    [Abstract | Paper.pdf]


      Submitted Papers

      Response adaptive designs for balancing complex objectives (with Q.F. Stout).
      Keywords: controlled clinical trial, Gittins index, multiple objectives, dynamic programming, bandit problem, optimal tradeoff, multiple objectives, modified bandit design     abstract   paper.pdf


      Conference Proceedings

      Recent progress in repeated significance tests, (1989) American Statistical Association: Proceedings of the Biopharmaceutical Section, 9-15.

      Supercomputer-intensive multivariable randomization tests, (with N. Schork) (1990). Computing Science and Statistics, 22: 509- 513. Wiley, New York.

      Bandit strategies for ethical sequential allocation, (with Q. F. Stout) (1991). Computing Science and Statistics, 23: Wiley, New York.
      Keywords: sequential allocation, dynamic programming, ethics, Gittens index, power, probability of correct selection, indifference region. [Abstract | Paper (Postscript) | Paper (PDF)]

      Optimal allocation for estimating the product of two means, (with Q. F. Stout) (1992). Computing Science and Statistics, 24: 592-596.
      Keywords: sequential allocation, nonlinear estimation, dynamic programming, reliability, myopic allocation, fixed allocation. [Abstract | Paper (Postscript) | Paper (PDF)]

      Optimal adaptive equal allocation rules, (with Q. F. Stout) (1992). Computing Science and Statistics, 24: 597-601.
      Keywords: sequential allocation, dynamic programming, curtailment, probability of correct selection, equal allocation, vector at a time. [Abstract | Paper (Postscript) | Paper (PDF)]

      Analysis of communication times on the IBM SP2, (with T. Tabe and Q. F. Stout) (1995). Computing Science and Statistics, 27, 347-351.
      Keywords: all-to-all communication, resampling, performance evaluation, message passing, operating systems, interrupts, heavy tails [Abstract | Paper (Postscript) | Paper (PDF)]

      Determining optimal few-stage allocation rules, (with Q. F. Stout) (1995). Computing Science and Statistics, 27, 342-346.
      93 (1998), pp. 1502-1511. Keywords: sequential allocation, selection, estimation, dynamic programming. [Abstract | Paper (Postscript) | Paper (PDF)]

      Sequential allocation with minimal switching, (with Q. F. Stout) (1996). Computing Science and Statistics, 28, 567-572.
      Keywords: adaptive sampling, switching costs, constraints, bandit, estimation, dynamic programming, optimal tradeoffs. [Abstract | Paper (Postscript) | Paper (PDF)]

      A parallel program for 3-arm bandits, (with R. Oehmke and Q. F. Stout) (1997). Computing Science and Statistics, 29, 390-395.
      Keywords: adaptive sampling, bandit, estimation, dynamic programming, forward induction, parallel computing. [Abstract | Paper (Postscript) | Paper (PDF)]

      Predicting algorithm performance, (with J. Landrum and Q. F. Stout) (1998). Computing Science and Statistics, 30, pp. 309-314.
      [Abstract | Paper (Postscript) | Paper (PDF)]

      Adaptive allocation with censoring, (with R. Oehmke and Q. F. Stout) (1998). Computing Science and Statistics, 30, pp. 219-223.
      [Abstract | Paper (Postscript) | Paper (PDF)]

      Minimizing the costs of screening trials,(with Q. F. Stout) (1999). Computing Science and Statistics 31.
      Keywords: few-stage, stopping rule. [Abstract | Paper (PDF) ]

      Optimal screening designs with flexible cost structures (with Q.F. Stout), Simulation 2001, S.M. Ermakov, Yu.N. Kashtanov, and V.B. Melas, eds., NII Chemistry St. Petersburg, 2001, pp. 253-260.
      Keywords: drug testing, acceptance-rejection sampling, few-stage, stopping rule, dynamic programming, design of experiments, decision theory. [Abstract | Paper (PDF)]

        jphard@umich.edu
        Last modified: December 2007