Abstract
In maximum entropy sampling (MES), a design is chosen by maximizing the joint Shannon entropy of parameters and observations. However, when the conditional parametric model of the response contains a large number of covariates, the posterior calculations in MES can be challenging or infeasible. In this work, we consider the use of composite likelihood modelling to break down the complexity of the full likelihood and code the original optimization problem into a set of simple partial likelihood problems. We study the optimality behaviour of the composite likelihood sampling approach as the number of design variables grows using both asymptotic analysis and numerical simulations.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Cox, D.R., Reid, N.: A note on pseudolikelihood constructed from marginal densities. Biometrika 91, 729–737 (2004)
Goldberg, D.E.: Genetic Algorithms in Search, Optimization and Machine Learning. Kluwer Academic, Boston (1989)
Johnson, M.E., Moore, L.M., Ylvisaker, D.: Minimax and maximin distance design. J. Stat. Plan. Inference 26, 131–148 (1990)
Ko, C.W., Lee, J., Queyranne, M.: An exact algorithm for maximum entropy sampling. Oper. Res. 43, 684–691 (1995)
Lindsey, D.V.: On a measure of the information provided by an experiment. Ann. Math. Stat. 27, 986–1005 (1956)
Lindsey, B.G., Yi, G.Y., Sum, J.: Issues and strategies in the selection of composite likelihoods. Stat. Sin. 21, 71–105 (2011)
Sebastiani, P., Wynn, H.P.: Maximum entropy sampling and optimal Bayesian experimental design. J. R. Stat. Soc. B 62, 145–157 (2000)
Singer, A.: Maximum entropy formulation of the Kirkwood superposition approximation. J. Chem. Phys. 121, 3657–3666 (2004)
Van der Vaart, A.W.: Asymptotic Statistics. Cambridge University Press, New York (1998)
Varin, C., Reid, N., Firth, D.: An overview of composite likelihood methods. Stat. Sin. 21, 5–42 (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
Proof of Proposition 1.
Let Z=(Z 1,…,Z p ) be a p-dimensional random vector with distribution p(z). Singer (2004) shows that if Z j is independent of Z k for any j≠k, then \(p(\mathbf{z}) = \tilde{p}(\mathbf{z})^{(p)} = \prod_{|E|<p} p_{E}(\mathbf{z}_{E})^{q_{E}}\), where E is a set in the power set of indexes , |E| is the cardinality of the set E, q E=(−1)p+1−|E|, and p E denotes the distribution of Z E ⊂Z. Without loss of generality, we start from θ 1, θ 2 and θ 3 and write
Recursively applying the formula by Singer (2004) for 3≤k≤p, gives
By summing over all such decompositions and taking the expectation with respect to Θ|Y,ξ, we obtain
which implies \(L(\xi) = \operatorname {E}_{Y} H(\varTheta|Y, \xi) = L^{(p)}(\xi) + S_{p}(\varTheta|Y,\xi) \). Finally, by our sparsity assumption, the last summand converges to zero as p→∞.
Rights and permissions
Copyright information
© 2013 Springer International Publishing Switzerland
About this paper
Cite this paper
Ferrari, D., Borrotti, M. (2013). Maximum Entropy Design in High Dimensions by Composite Likelihood Modelling. In: Ucinski, D., Atkinson, A., Patan, M. (eds) mODa 10 – Advances in Model-Oriented Design and Analysis. Contributions to Statistics. Springer, Heidelberg. https://doi.org/10.1007/978-3-319-00218-7_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-00218-7_9
Publisher Name: Springer, Heidelberg
Print ISBN: 978-3-319-00217-0
Online ISBN: 978-3-319-00218-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)