Abstract
This paper presents the link between stochastic approximation and multi-arm clinical trials based on randomized urn models investigated in Bai et al. (J. Multivar. Anal. 81(1):1–18, 2002) where the urn updating depends on the past performances of the treatments. We reformulate the dynamics of the urn composition, the assigned treatments and the successes of assigned treatments as standard stochastic approximation (SA) algorithms with remainder. Then, we derive the a.s. convergence of the normalized procedure under less stringent assumptions by calling upon the ODE and a new asymptotic normality result (Central Limit Theorem CLT) by calling upon the SDE methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Andersena J, Fariesa D, Ramuraa R (1994) A randomized play-the-winner design for multi-arm clinical trials. Commun Stat, Theory Methods 23:309–323
Athreya KB, Karlin S (1967) Limit theorems for the split times of branching processes. J Math Mech 17:257–277
Athreya KB, Karlin S (1968) Embedding of urn schemes into continuous time Markov branching processes and related limit theorems. Ann Math Stat 39:1801–1817
Bai Z-D, Hu F (1999) Asymptotic theorems for urn models with nonhomogeneous generating matrices. Stoch Process Appl 80(1):87–101
Bai Z-D, Hu F (2005) Asymptotics in randomized urn models. Ann Appl Probab 15(1B):914–940
Bai Z-D, Hu F, Shen L (2002) An adaptive design for multi-arm clinical trials. J Multivar Anal 81(1):1–18
Benaïm M (1999) Dynamics of stochastic approximation algorithms. In: Séminaire de probabilités, XXXIII. Lecture notes in mathematics, vol 1709. Springer, Berlin, pp 1–68
Benveniste A, Métivier M, Priouret P (1990) Adaptive algorithms and stochastic approximations. Applications of mathematics (New York), vol 22. Springer, Berlin. Translated from the French by Stephen S. Wilson
Berman A, Plemmons RJ (1994) Nonnegative matrices in the mathematical sciences. Classics in applied mathematics, vol 9. SIAM, Philadelphia. Revised reprint of the 1979 original
Duflo M (1997) Random iterative models. Applications of mathematics (New York), vol 34. Springer, Berlin. Translated from the 1990 French original by Stephen S. Wilson and revised by the author
Flournoy N, Rosenberger WF (eds) (1995) Adaptive designs. Institute of mathematical statistics lecture notes—monograph series, vol 25. Institute of Mathematical Statistics, Hayward. Papers from the Joint AMS-IMS-SIAM Summer conference held at Mt, Holyoke College, South Hadley, MA, 1992
Fort J-C, Pagès G (1996) Convergence of stochastic algorithms: from the Kushner-Clark theorem to the Lyapounov functional method. Adv Appl Probab 28(4):1072–1094
Freedman DA (1965) Bernard Friedman’s urn. Ann Math Stat 36:956–970
Gouet R (1993) Martingale functional central limit theorems for a generalized Pólya urn. Ann Probab 21(3):1624–1639
Janson S (2004) Functional limit theorems for multitype branching processes and generalized Pólya urns. Stoch Process Appl 110(2):177–245
Johnson NL, Kotz S (1977) Urn models and their application. Wiley, New York
Kushner HJ, Yin GG (2003) Stochastic approximation and recursive algorithms and applications 2nd edn. Applications of Mathematics (New York), vol 35. Springer, New York
Laruelle S, Pagès G (2013) Randomized urn models revisited using stochastic approximation. Ann Appl Probab 23(4):1409–1436. doi:10.1214/12-AAP875. Permanent link to this document: http://projecteuclid.org/euclid.aoap/1371834033
Pouyanne N (2008) An algebraic approach to Pólya processes. Ann Inst Henri Poincaré 44(2):293–323
Rosenberger WF (1996) New directions in adaptive designs. Stat Sci 11(2):137–149
Smythe RT (1996) Central limit theorems for urn models. Stoch Process Appl 65(1):115–137
Wei LJ (1979) The generalized Polya’s urn design for sequential medical trials. Ann Stat 7(2):291–296
Zelen M (1969) Play the winner rule and the controlled clinical trial. J Am Stat Assoc 64:131–146
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix: Basic Tools of Stochastic Approximation
Appendix: Basic Tools of Stochastic Approximation
Consider the following recursive procedure defined on a filtered probability space \((\varOmega,{\mathcal{A}},({\mathcal{F}}_{n})_{n\geq0},{\mathbb{P}})\)
where \(h:{\mathbb{R}}^{d}\rightarrow {\mathbb{R}}^{d}\) is a locally Lipschitz continuous function, \(\theta_{n_{0}}\) an \({\mathcal{F}}_{n_{0}}\)-measurable finite random vector and, for every n≥n 0, ΔM n+1 is an \({\mathcal{F}}_{n}\)-martingale increment and r n is an \({\mathcal{F}}_{n}\)-adapted remainder term.
Theorem 3.3
(A.s. Convergence with ODE Method, see e.g. [7, 8, 10, 12, 17])
Assume that h is locally Lipschitz, that
and that (γ n ) n≥1 is a positive sequence satisfying
Then the set Θ ∞ of its limiting values as n→+∞ is a.s. a compact connected set, stable by the flow of
Furthermore if θ ∗∈Θ ∞ is a uniformly stable equilibrium on Θ ∞ of ODE h , then
Comments
By uniformly stable we mean that
where \(\theta(\theta_{0},t)_{\theta_{0}\in\varTheta^{\infty},t\in {\mathbb{R}}_{+}}\) is the flow of ODE h on Θ ∞.
Theorem 3.4
(Rate of Convergence see [10], Theorem 3.III.14, p.131 (for CLT see also e.g. [8, 17]))
Let θ ∗ be an equilibrium point of {h=0}. Assume that the function h is differentiable at θ ∗ and all the eigenvalues of Dh(θ ∗) have positive real parts. Assume that for some δ>0,
where Γ is a deterministic symmetric definite positive matrix and for an ϵ>0,
Specify the gain parameter sequence as follows: for every n≥1, \(\gamma_{n}=\frac{1}{n}\). If \(\varLambda:=\Re e(\lambda_{{\rm min}})>\frac {1}{2}\), where \(\lambda_{{\rm min}}\) denotes the eigenvalue of Dh(θ ∗) with the lowest real part, then, the above a.s. convergence is ruled on the convergence set {θ n →θ ∗} by the following Central Limit Theorem
with
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Laruelle, S., Pagès, G. (2014). Urn Model-Based Adaptive Multi-arm Clinical Trials: A Stochastic Approximation Approach. In: Abergel, F., Aoyama, H., Chakrabarti, B., Chakraborti, A., Ghosh, A. (eds) Econophysics of Agent-Based Models. New Economic Windows. Springer, Cham. https://doi.org/10.1007/978-3-319-00023-7_3
Download citation
DOI: https://doi.org/10.1007/978-3-319-00023-7_3
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-00022-0
Online ISBN: 978-3-319-00023-7
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)