Electronic Journal of Statistics

Statistical properties of two-color randomly reinforced urn design targeting fixed allocations

Abstract

This paper deals with the statistical properties of a response adaptive design, described in terms of a two colors urn model, targeting prespecified asymptotic allocations. Results on the rate of divergence of number of patients assigned to each treatment are proved as well as on the asymptotic behavior of the urn composition. Suitable statistics are introduced and studied to test the hypothesis on treatments’ difference.

Article information

Source
Electron. J. Statist., Volume 8, Number 1 (2014), 708-737.

Dates
First available in Project Euclid: 21 May 2014

https://projecteuclid.org/euclid.ejs/1400703411

Digital Object Identifier
doi:10.1214/14-EJS899

Mathematical Reviews number (MathSciNet)
MR3211029

Zentralblatt MATH identifier
1348.62254

Citation

Ghiglietti, Andrea; Paganoni, Anna Maria. Statistical properties of two-color randomly reinforced urn design targeting fixed allocations. Electron. J. Statist. 8 (2014), no. 1, 708--737. doi:10.1214/14-EJS899. https://projecteuclid.org/euclid.ejs/1400703411

References

• [1] Aletti, G., May, C. and Secchi, P. (2007). On the distribution of the limit proportion for a two-color, randomly reinforced urn with equal reinforcement distributions. Advances in Applied Probability, 39, 690–707.
• [2] Aletti, G., May, C. and Secchi, P. (2009a). A Central Limit Theorem, and related results, for two-color randomly reinforced urn. Advances in Applied Probability, 41, 829–844.
• [3] Aletti, G., May, C. and Secchi, P. (2012). A functional equation whose unknown is $P([0;1])$ valued. Journal of Theoretical Probability, 25(4), 1207–1232.
• [4] Aletti, G., Ghiglietti, A. and Paganoni, A. M. (2013). A modified randomly reinforced urn design. Forthcoming in: Journal of Applied Probability, 50(2).
• [5] Bai, Z. D., Hu, F. and Zhang, L. X. (2002). Gaussian approximation theorems for urn models and their applications. Annals of Applied Probabability, 12, 1149–1173.
• [6] Bai, Z. D. and Hu, F. (2005). Asymptotics in randomized urn models. Annals of Applied Probabability, 15, 914–940.
• [7] Beggs, A. W. (2005). On the convergence of reinforcement learning. Journal of Economic Theory, 122, 1–36.
• [8] Cheung, Z. D., Hu, F. and Zhang, L. X. (2006). Asymptotic theorems of sequential estimation-adjusted urn models. Annals of Applied Probabability, 16, 340–369.
• [9] Cheung, Z. D., Chan, W. S., Hu, F. and Zhang, L. X. (2011). Immigrated Urn Models – theoretical properties and applications. Annals of Statistics, 39(1), 643–671.
• [10] Durham, S. C. and Yu, K. F. (1990). Randomized play-the leader rules for sequential sampling from two populations. Probability in Engineering and Information Science, 26(4), 355–367.
• [11] Durham, S. C., Flournoy, N. and Li, W. (1996). Randomized Pólya urn designs. Proc. Biometric Section of the American Statistical Association, pp. 166–170. Amer. Statist. Assoc., Alexandria, VA.
• [12] Durham, S. C., Flournoy, N. and Li, W. (1998). A sequential design for maximizing the probability of a response. The Canadian Journal of Statistics, 26(3), 479–495.
• [13] Flournoy, N. and May, C. (2009). Asymptotics in response-adaptive designs generated by a two-color, randomly reinforced urn. Annals of Statistics, 37, 1058–1078.
• [14] Flournoy, N., May, C., Moler, J. A. and Plo, F. (2009). On testing hypotheses in response-adaptive designs targeting the best treatment. In mODa 9-Advances in Model-Oriented Design and Analysis, Contrib. Statist., pp. 81–88. Physica-Verlag/Springer, Heidelberg.
• [15] Flournoy, N., May, C. and Secchi, P. (2012). Asymptotically optimal response-adaptive designs for allocating the best treatment: An overview. International Statistical Review, 80(2), 293–305.
• [16] Ghiglietti, A. and Paganoni, A. M. (2013). An urn procedure for efficiently detecting the parameters of an hypothesis test. Available as Mox-report/2013, Politecnico di Milano. In: http://mox.polimi.it/it/progetti/pubblicazioni/quaderni/60-2013.pdf.
• [17] Hu, F. and Rosenberger, W. F. (2003). Optimality, variability, power: Evaluating responseadaptive randomization procedures for treatment comparisons. Journal of American Statistical Society, 98, 671–678.
• [18] Hu, F. and Rosenberger, W. F. (2006). The Theory of Response-Adaptive Randomization in Clinical Trials. Wiley, Hoboken, NJ.
• [19] Huang, P. (2007). A numerical study of a randomized reinforced urn design for binary responses. Technical report, Univ. Missouri-Columbia. A Masters’ project available from N. Flournoy upon request.
• [20] Janson, S. (2004). Functional limit theorems for multitype branching processes and generalized Pólya urns. Stochastic Process. Appl., 110, 177–245.
• [21] Lachin, J. M. and Rosenberger, W. F. (2006). Randomization in Clinical Trials. Wiley, New York.
• [22] May, C., Paganoni, A. M. and Secchi, P. (2005). On a two-color generalized Polya urn. Metron, Vol. LXIII, 1.
• [23] May, C., Paganoni, A. M. and Secchi, P. (2007). Response-adaptive designs targeting the best treatment for clinical trials with continuous responses. In S. Co. 2007 Fifth Conference-Complex Models and Computational Intensive Methods for Estimation and Prediction-Book of Short Papers, pp. 326–331. Cleup, Padova.
• [24] Melfi, F. and Page, C. (2000). Estimation after adaptive allocation. J. Statist. Plann. Inference, 87, 353–363.
• [25] Melfi, F., Page, C. and Geraldes, M. (2001). An adaptive randomized design with application to estimation. The Canadian Journal of Statistics, 29, 107–116.
• [26] Muliere, P., Paganoni, A. M. and Secchi, P. (2006). A randomly reinforced urn. Journal of Statistical Planning and Inference, 136, 1853–1874.
• [27] Paganoni, A. M. and Secchi, P. (2007). A numerical study for comparing two response-adaptive designs for continuous treatment effects. Statistical Methods and Applications, 16, 321–346.
• [28] Rosenberger, W. F. (2002). Randomized urn models and sequential design. Sequential Analysis, 21(1&2), 1–28.
• [29] Rosenberger, W. F. and Zhang, L. (2006). Response-adaptive randomization for clinical trials with continuous responses. Biometrics, 62, 562–569.