We present a new response adaptive design, described in terms of a two colors urn model with non-negative and diagonal reinforcement matrix. The model consists in a modification of the Randomly Reinforced Urn design aiming to target any prespecified asymptotic allocation. Besides the main convergence theorem, we present further asymptotic results concerning the asymptotic behaviour of different quantities related to the urn process. We introduce suitable statistics to test the hypothesis on treatment’s differences and we study the statistical proprieties of these inferential procedures.

Ghiglietti, A., Paganoni, A. (2013). Asymptotic statistical properties of a response adaptive design based on a two colours urn model. In Proceedings of SCo 2013 - 8th Conference on Complex Data Modelling and Computationally Intensive Statistical Methods for Estimation and Prediction (pp.1-6). DEU : Spring.

Asymptotic statistical properties of a response adaptive design based on a two colours urn model

Ghiglietti, A;
2013

Abstract

We present a new response adaptive design, described in terms of a two colors urn model with non-negative and diagonal reinforcement matrix. The model consists in a modification of the Randomly Reinforced Urn design aiming to target any prespecified asymptotic allocation. Besides the main convergence theorem, we present further asymptotic results concerning the asymptotic behaviour of different quantities related to the urn process. We introduce suitable statistics to test the hypothesis on treatment’s differences and we study the statistical proprieties of these inferential procedures.
No
paper
Scientifica
response adaptive design, two colors urn model
English
S.Co. 2013 - Complex Data Modeling and Computational Intensive Statistical Methods for Computation and Prediction
9788864930190
Ghiglietti, A., Paganoni, A. (2013). Asymptotic statistical properties of a response adaptive design based on a two colours urn model. In Proceedings of SCo 2013 - 8th Conference on Complex Data Modelling and Computationally Intensive Statistical Methods for Estimation and Prediction (pp.1-6). DEU : Spring.
Ghiglietti, A; Paganoni, A
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/10281/391740
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact