| P. Sebastiani and H. P. Wynn. Maximum entropy sampling and optimal bayesian experimental design. J. Roy. Stat. Soc. B, 62:145--157, 2000. |
....is small. Needless to say that this is the case in which an optimal solution is mostly required. In this paper we show that, in some cases, there exists a dual solution to the design problem that partly overcomes computational difficulties. Bayesian Experimental Design 2 The approach is due to Sebastiani and Wynn (1997) and extends a method described in Shewry and Wynn (1987) for predictive design problems to estimative design problems. The main result is that, under general assumptions, the minimization of the expected posterior entropy of Theta w.r.t. the experiment is achieved by maximizing the marginal ....
....the observations. This yields an alternative design criterion called Maximum Entropy Sampling (mes) In Section 2, we describe the theory that yields the formulation of the dual design criterion. Computational advantages are discussed in Section 3, and an example is given in Section 4 2. Theory Sebastiani and Wynn (1997), introduce mes, for estimation problems. The principle is that if the entropy of the sampling distribution is not functionally dependent on the design, then a design minimizing the expected posterior entropy can be found by maximizing the marginal entropy of the data. The result is obtained by ....
[Article contains additional citation context not shown here]
Sebastiani, P., & Wynn, H. P. (1997). Maximum entropy sampling and optimal Bayesian experimental design. J.Roy.Statist.Soc.B, To appear.
No context found.
P. Sebastiani and H. P. Wynn. Maximum entropy sampling and optimal bayesian experimental design. J. Roy. Stat. Soc. B, 62:145--157, 2000.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC