Abstract
In nonlinear regression, optimal designs of experiments generally depend on unknown parameters. One approach to deal with this problem is to use the minimax criterion for choosing a design. However, constructing minimax designs has shown to be numerically intractable in many applications. The H-algorithm is a new iterative algorithm that utilizes the relation between minimax designs and optimum on-the-average designs based on a least favorable distribution. It is also fairly easy to apply this algorithm in applications. In this paper, the H-algorithm is described and its convergence properties are discussed.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Atkinson, A.C., Donev, A.N., Tobias, R.D.: Optimum Experimental Designs, with SAS. Oxford University Press, New York (2007)
Berger, M.P.F., Wong, W.K.: An Introduction to Optimal Designs for Social and Biomedical Research. Wiley, Chichester (2009)
Dette, H., Martinez Lopez, I., Ortiz Rodriguez, I.M., Pepelyshev, A.: Maximin efficient design of experiment for exponential regression models. J. Stat. Plan. Inference 136, 4397–4418 (2006)
Fackle-Fornius, E., Nyquist, H.: Optimal allocation to treatment groups under variance heterogeneity. Research Report, Department of Statistics, Stockholm University (2012)
Fackle-Fornius, E., Wänström, L.: Minimax optimal designs of contingent valuation experiment: willingness to pay for environmentally friendly clothes. Working paper, Department of Statistics, Stockholm University (2012)
Fedorov, V.V., Hackl, P.: Model-Oriented Design of Experiments. Springer, New York (1997)
Imhof, L.A.: Maximin designs for exponential growth models and heteroscedastic polynomial models. Ann. Stat. 29, 561–576 (2001)
Kiefer, J.: General equivalence theory for optimum designs (approximate theory). Ann. Stat. 2, 849–879 (1974)
King, J., Wong, W.K.: Optimal minimax designs for prediction in heteroscedastic models. J. Stat. Comput. Simul. 69, 371–383 (1998)
King, J., Wong, W.K.: Optimal designs for the power logistic model. J. Stat. Comput. Simul. 74, 779–791 (2004)
Müller, C.: Maximin efficient designs for estimating nonlinear aspects in linear models. J. Stat. Plan. Inference 44, 117–132 (1995)
Müller, C., Pazman, A.: Applications of necessary and sufficient conditions for maximin efficient designs. Metrika 48, 1–19 (1998)
Noubiap, R.F., Seidel, W.: A minimax algorithm for constructing optimal symmetrical balanced designs for a logistic regression model. J. Stat. Plan. Inference 91, 151–168 (2000)
Nyquist, H.: Conditions for minimax designs. Working paper, Stockholm University (2012)
Sitter, R.: Robust designs for binary data. Biometrics 48, 1145–1155 (1992)
Acknowledgements
This work was supported by a grant from the Swedish Research Council.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer International Publishing Switzerland
About this paper
Cite this paper
Nyquist, H. (2013). Convergence of an Algorithm for Constructing Minimax Designs. 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_22
Download citation
DOI: https://doi.org/10.1007/978-3-319-00218-7_22
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)