×

Optimal designs for experiments with a large number of factors. (English) Zbl 0832.62070

Summary: Optimal designs are considered for experiments with many binary factors. In the model, only main effects are considered and responses are normal with equal variances. The design objective is to determine factor settings that give a large expected response after experimentation. Asymptotically optimal designs are obtained using a Bayesian decision theoretic formulation in which factors are independent with finite Fisher information. Both the number of runs and the number of factors tend to infinity in the limit considered. In the derivation, Stein’s identity provides accurate approximations for posterior risks. The optimal number of runs for experiments with many factors is determined.

MSC:

62K05 Optimal statistical designs
62C10 Bayesian problems; characterization of Bayes procedures
Full Text: DOI

References:

[1] Baumer, L. D.; Golomb, S. W.; Hall, M. J., Discovery of an Hadamard matrix of order 92, Bull. Amer. Math. Soc., 68, 237-238 (1962) · Zbl 0105.01002
[2] Berger, J. O., Statistical Decision Theory and Bayesian Analysis (1985), Springer: Springer New York · Zbl 0572.62008
[3] Bickel, P.; Goetze, F.; Van Zwet, W., A simple analysis of third order efficiency of estimates, (Proc. Neyman-Kiefer Conf., Vol. 2 (1985)), 749-768 · Zbl 1373.62093
[4] Chernoff, H., Sequential Analysis and Optimal Design (1972), SIAM: SIAM Philadelphia · Zbl 0265.62024
[5] Fedorov, V. V., Theory of Optimal Experiments (1972), Academic Press: Academic Press New York
[6] Ghosh, B. K.; Sen, P. K., (Handbook of Sequential Analysis (1991), Marcel Dekker: Marcel Dekker New York) · Zbl 0753.62046
[7] Ghosh, J. K.; Sinha, B.; Joshi, S., Expansions for posterior distributions and integrated Bayes risk, (Proc. 3rd Purdue Symp. Statist. Dec. Th. Rel. Topics, Vol. 1 (1985)), 403-456
[8] Johnson, R. A., Asymptotic expansions associated with posterior distributions, Ann. Statist., 41, 851-864 (1970) · Zbl 0204.53002
[9] Keener, R.; Woodroofe, M., Fixed width interval estimation for the reciprocal drift of Brownian motion, J. Statist. Plann. Inference, 30, 1-12 (1992) · Zbl 0766.62052
[10] Pilz, J., Bayesian Estimation and Experimental Design in Linear Regression Models (1991), Wiley: Wiley New York · Zbl 0745.62068
[11] Plackett, R. L.; Burman, J. P., The designs of optimum multifactorial experiments, Biometrika, 33, 305-325 (1946) · Zbl 0063.06274
[12] Silvey, S. D., Optimal Design (1980), Chapman and Hall: Chapman and Hall London · Zbl 0468.62070
[13] Stein, C., Approximate Computation of Expectations (1986), IMS: IMS California · Zbl 0721.60016
[14] Woodroofe, M., Integrable expansions for posterior distributions for one-parameter exponential families, Statistica Sinica, 2, 91-111 (1992) · Zbl 0820.62019
[15] Woodroofe, M.; Hardwick, J., Sequential allocation for an estimation problem with ethical costs, Ann. Statist., 18, 1358-1377 (1990) · Zbl 0712.62074
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.