Skip to main content

An urn model to construct an efficient test procedure for response adaptive designs

Abstract

We study the statistical performance of different tests for comparing the mean effect of two treatments. Given a reference classical test \({\mathcal {T}}_0\), we determine which sample size and proportion allocation guarantee to a test \({\mathcal {T}}\), based on response-adaptive design, to be better than \({\mathcal {T}}_0\), in terms of (a) higher power and (b) fewer subjects assigned to the inferior treatment. The adoption of a response-adaptive design to implement the random allocation procedure is necessary to ensure that both (a) and (b) are satisfied. In particular, we propose to use a Modified Randomly Reinforced Urn design and we show how to perform the model parameters selection for the purpose of this paper. Then, the opportunity of relaxing some assumptions on treatment response distributions is presented. Results of simulation studies on the test performance are reported and a real case study is analyzed.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

References

  1. Aletti G, Ghiglietti A, Paganoni AM (2013) Randomly reinforced urn designs with prespecified allocations. J Appl Probab 50(2):486–498

    MathSciNet  Article  MATH  Google Scholar 

  2. Atkinson F, Biswas A (2014) Randomised response-adaptive designs in clinical trials. CRC Press, Boca-Raton

    MATH  Google Scholar 

  3. Bai ZD, Hu F, Zhang LX (2002) Gaussian approximation theorems for urn models and their applications. Ann Appl Probab 12:1149–1173

    MathSciNet  Article  MATH  Google Scholar 

  4. Bai ZD, Hu F (2005) Asymptotics in randomized urn models. Ann Appl Probab 15:914–940

    MathSciNet  Article  MATH  Google Scholar 

  5. Beggs AW (2005) On the convergence of reinforcement learning. J Econ Theory 122:1–36

    MathSciNet  Article  MATH  Google Scholar 

  6. Cheung ZD, Hu F, Zhang LX (2006) Asymptotic theorems of sequential estimation-adjusted urn models. Ann Appl Probab 16:340–369

    MathSciNet  Article  MATH  Google Scholar 

  7. Cheung ZD, Chan WS, Hu F, Zhang LX (2011) Immigrated urn models—theoretical properties and applications. Ann Stat 39(1):643–671

    MathSciNet  Article  MATH  Google Scholar 

  8. Durham SC, Yu KF (1990) Randomized play-the leader rules for sequential sampling from two populations. Probab Eng Inf Sci 26(4):355–367

    Article  MATH  Google Scholar 

  9. Durham SC, Flournoy N, Li W (1996) Randomized plya urn designs. In: Proceedings Biometric Section of the American Statistical Association. Am Stat Assoc, Alexandria, VA, pp 166–170

  10. Durham SC, Flournoy N, Li W (1998) A sequential design for maximizing the probability of a response. Can J Stat 26(3):479–495

    MathSciNet  Article  MATH  Google Scholar 

  11. Flournoy N, May C (2009) Asymptotics in response-adaptive designs generated by a two-color, randomly reinforced urn. Ann Stat 37:1058–1078

    MathSciNet  Article  MATH  Google Scholar 

  12. Flournoy N, May C, Secchi P (2012) Asymptotically optimal response-adaptive designs for allocating the best treatment: an overview. Int Stat Rev 80(2):293–305

    MathSciNet  Article  Google Scholar 

  13. Ghiglietti A, Paganoni AM (2014) Statistical properties of two-color randomly reinforced urn design targeting fixed allocations. Electron J Stat 8(1):708-170

    MathSciNet  Article  MATH  Google Scholar 

  14. Ghiglietti A (2014) Statistical properties of urn models in response-adaptive designs. PhD thesis. Politecnico di Milano

  15. Grieco N, Ieva F, Paganoni AM (2012) Performance assessment using mixed effects models: a case study on coronary patient care. IMA J Manag Math 23:117–131

    Article  MATH  Google Scholar 

  16. Hu F, Rosenberger WF (2006) The theory of response-adaptive randomization in clinical trials. Wiley, Hoboken, NJ

    Book  MATH  Google Scholar 

  17. Janson S (2004) Functional limit theorems for multitype branching processes and generalized pólya urns. Stoch Process Appl 110:177–245

    MathSciNet  Article  MATH  Google Scholar 

  18. Lachin JM, Rosenberger WF (2002) Randomization in clinical trials. Wiley, New York

    MATH  Google Scholar 

  19. Lehmann EL, Romano JP (2005) Testing statistical hypotheses. Springer, New York

    MATH  Google Scholar 

  20. Laruelle S, Pags G (2013) Randomized urn models revisited using stochastic approximation. Ann Appl Probab 23(4):1409–1436

    MathSciNet  Article  MATH  Google Scholar 

  21. May C, Paganoni AM, Secchi P (2005) On a two-color generalized Polya urn. Metron, vol. LXIII. p 1

  22. Melfi F, Page C (2000) Estimation after adaptive allocation. J Stat Plan Inference 87:353–363

    MathSciNet  Article  MATH  Google Scholar 

  23. Melfi F, Page C, Geraldes M (2001) An adaptive randomized design with application to estimation. Can J Stat 29:107–116

    MathSciNet  Article  MATH  Google Scholar 

  24. Muliere P, Paganoni AM, Secchi P (2006) A randomly reinforced urn. J Stat Plan Inference 136:1853–1874

    MathSciNet  Article  MATH  Google Scholar 

  25. Paganoni AM, Secchi P (2007) A numerical study for comparing two response-adaptive designs for continuous treatment effects. Stat Methods Appl 16:321–346

    MathSciNet  Article  MATH  Google Scholar 

  26. R Development Core Team (2011). A language and environment for statistical computing. R foundation for statistical computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/

  27. Rosenberger WF, Stallard N, Ivanova A, Harper CN, Ricks ML (2001) Optimal adaptive designs for binary response trials. Biometrics 57:909–913

    MathSciNet  Article  MATH  Google Scholar 

  28. Rosenberger WF (2002) Randomized urn models and sequential design. Seq Anal 21(1&2):1–28

    MathSciNet  Article  MATH  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Andrea Ghiglietti.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ghiglietti, A., Paganoni, A.M. An urn model to construct an efficient test procedure for response adaptive designs. Stat Methods Appl 25, 211–226 (2016). https://doi.org/10.1007/s10260-015-0314-y

Download citation

Keywords

  • Response adaptive designs
  • Clinical trials
  • Randomly reinforced urns
  • Tests based on adaptive procedures