×

A semi-infinite programming based algorithm for determining T-optimum designs for model discrimination. (English) Zbl 1314.62164

Summary: T-optimum designs for model discrimination are notoriously difficult to find because of the computational difficulty involved in solving an optimization problem that involves two layers of optimization. Only a handful of analytical T-optimal designs are available for the simplest problems; the rest in the literature are found using specialized numerical procedures for a specific problem. We propose a potentially more systematic and general way for finding T-optimal designs using a Semi-Infinite Programming (SIP) approach. The strategy requires that we first reformulate the original minimax or maximin optimization problem into an equivalent semi-infinite program and solve it using an exchange-based method where lower and upper bounds produced by solving the outer and the inner programs, are iterated to convergence. A global Nonlinear Programming (NLP) solver is used to handle the subproblems, thus finding the optimal design and the least favorable parametric configuration that minimizes the residual sum of squares from the alternative or test models. We also use a nonlinear program to check the global optimality of the SIP-generated design and automate the construction of globally optimal designs. The algorithm is successfully used to produce results that coincide with several T-optimal designs reported in the literature for various types of model discrimination problems with normally distributed errors. However, our method is more general, merely requiring that the parameters of the model be estimated by a numerical optimization.

MSC:

62K05 Optimal statistical designs
90C22 Semidefinite programming

Software:

GAMS; CONOPT
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Atkinson, A. C., The non-uniqueness of some designs for discriminating between two polynomial models in one variable, (Giovagnoli, A.; Atkinson, A. C.; Torsney, B., moDA 9 - Advances in Model-Oriented and Analysis, Proceedings of the 9th International Workshop in Model-Oriented Design and Analysis (2010), Springer-Verlag: Springer-Verlag Heidelberg), 9-16
[2] Atkinson, A. C., Optimum experimental designs for choosing between competitive and non competitive models for enzyme inhibition, Comm. Statist. Theory Methods, 41, 2283-2296 (2012) · Zbl 1271.62168
[3] Atkinson, A. C.; Cox, D. R., Planning experiments for discrimination between models (with discussion), J. R. Stat. Soc. Ser. B, 36, 321-348 (1974) · Zbl 0291.62095
[4] Atkinson, A. C.; Donev, A. N.; Tobias, R. D., Optimum Experimental Designs, with SAS (2007), Oxford University Press: Oxford University Press New York · Zbl 1183.62129
[5] Atkinson, A. C.; Fedorov, V. V., The design of experiments for discriminating between two rival models, Biometrika, 62, 57-70 (1975) · Zbl 0308.62071
[6] Atkinson, A. C.; Fedorov, V. V., Optimal design: experiments for discriminating between several models, Biometrika, 62, 289-303 (1975) · Zbl 0321.62085
[7] Berger, M. P.F.; Wong, W. K., (An Introduction to Optimal Designs for Social and Biomedical Research. An Introduction to Optimal Designs for Social and Biomedical Research, Statistics in Practice (2009), Wiley: Wiley New York) · Zbl 1182.62154
[8] Blankenship, J. W.; Falk, J. E., Infinitely constrained optimization problems, J. Optim. Theory Appl., 19, 261-281 (1976) · Zbl 0307.90071
[9] Dette, H., Discrimination designs for polynomial regression on compact intervals, Ann. Statist., 22, 890-903 (1994) · Zbl 0806.62059
[10] Dette, H.; Haller, G., Optimal designs for the identification of the order of the Fourier regression, Ann. Statist., 26, 196-1521 (1998) · Zbl 0930.62075
[11] Dette, H.; Melas, V. B.; Shpilev, P., T-optimal designs for discrimination between two polynomial models, Ann. Statist., 40, 188-205 (2012) · Zbl 1246.62176
[12] Dette, H.; Titoff, S., Optimal discriminating designs, Ann. Statist., 37, 2056-2081 (2009) · Zbl 1168.62066
[13] Drud, A., CONOPT: a GRG code for large sparse dynamic nonlinear optimization problems, Math. Program., 31, 153-191 (1985) · Zbl 0557.90088
[14] Duarte, B. P.M.; Wong, W. K., A Semi-Infinite Programming based algorithm for finding minimax optimal designs for nonlinear models, Stat. Comput., 24, 1063-1080 (2013) · Zbl 1332.90189
[15] Fedorov, V. V., Theory of Optimal Experiments (1972), Academic Press: Academic Press New York
[16] Fedorov, V. V.; Hackl, P., Model-oriented Design of Experiments (1997), Springer: Springer New York · Zbl 0878.62052
[17] Fedorov, V. V.; Leonov, S. L., Optimal Design for Nonlinear Response Models (2014), Chapman and Hall/ CRC Press: Chapman and Hall/ CRC Press Boca Raton · Zbl 1373.62001
[18] Fedorov, V. V.; Malyutov, M. B., Optimal designs in regression problems, Math. Oper.forsch. Stat., 3, 281-308 (1972) · Zbl 0258.62044
[20] Goos, P.; Jones, B., Optimal Design of Experiments: A Case Study Approach (2011), Wiley: Wiley New York
[21] Gribik, P. R.; Kortanek, K. O., Equivalence theorems and cutting plane algorithms for a class of experimental design problems, SIAM J. Appl. Math., 32, 232-259 (1977) · Zbl 0356.62060
[22] Hettich, R.; Kaplan, A.; Tichatschke, R., Semi-Infinite Programming: numerical methods, (Encyclopedia of Optimization, Vol. 5 (2001), Kluwer: Kluwer Amsterdam), 112-117
[23] Hettich, R.; Kortanek, K. O., Semi-infinite programming: theory, methods and applications, SIAM Rev., 35, 380-429 (1993) · Zbl 0784.90090
[24] Kiefer, J. C., Optimum experimental designs, J. R. Stat. Soc. Ser. B, 21, 272-319 (1959) · Zbl 0108.15303
[25] Kiefer, J., General equivalence theory for optimum design (approximate theory), Ann. Statist., 2, 849-879 (1974) · Zbl 0291.62093
[26] Kiefer, J.; Wolfowitz, J., The equivalence of two extremum problem, Canad. J. Math., 12, 363-366 (1960) · Zbl 0093.15602
[27] Kuczewski, B., Computational Aspects of Discrimination between Models of Dynamic Systems (2005), University of Zielona Góra: University of Zielona Góra Zielona Góra, Poland, (Ph.D. thesis)
[28] López, M.; Still, G., Semi-Infinite Programming, European J. Oper. Res., 180, 491-518 (2007) · Zbl 1124.90042
[29] López-Fidalgo, J.; Trandafir, C.; Tommasi, C., An optimal experimental design criterion for discriminating between non-normal models, J. R. Stat. Soc. Ser. B, 69, 231-242 (2007) · Zbl 1123.62056
[30] López-Fidalgo, J.; Tommasi, C.; Trandafir, C., Optimal designs for discriminating between some extensions of the Michaelis-Menten model, J. Statist. Plann. Inference, 138, 3797-3804 (2008) · Zbl 1146.62057
[31] Mitsos, A.; Lemonidis, P.; Barton, P. I., Global solution of bilevel programs with a nonconvex inner program, J. Global Optim., 42, 475-513 (2008) · Zbl 1163.90700
[32] Polak, E., Optimization: Algorithms and Consistent Approximations (1997), Springer: Springer New York · Zbl 0899.90148
[33] Ponce de Leon, A. C.; Atkinson, A. C., Optimum experimental design for discriminating between two rival models in the presence of prior information, Biometrika, 78, 601-608 (1991) · Zbl 0741.62073
[34] Pukelsheim, F., Optimal Design of Experiments (1993), SIAM: SIAM Philadelphia · Zbl 0834.62068
[35] Pukelsheim, F.; Rieder, S., Efficient rounding of approximate designs, Biometrika, 79, 763-770 (1992)
[36] Reemtsen, R.; Rückman, J. J., Semi-Infinite Programming (1998), Kluwer: Kluwer Amsterdam
[37] Rustem, B.; Howe, M., Algorithms for Worst-case Design and Applications to Risk Management (2002), Princeton University Press: Princeton University Press Princeton, NJ · Zbl 1140.90013
[38] Shimizu, K.; Aiyoshi, E., Necessary conditions for min-max problems and algorithms by a relaxation procedure, IEEE Trans. Automat. Control, 25, 62-66 (1980) · Zbl 0426.49008
[39] Skanda, D.; Lebiedz, D., A robust optimization approach to experimental design for model discrimination of dynamical systems, Math. Program. A, 1-29 (2012)
[40] Tsoukalas, A.; Rustem, B.; Pistikopoulos, E. N., A global optimization algorithm for generalized semi-infinite, continuous minimax with coupled constraints and bi-level problems, J. Global Optim., 24, 235-250 (2009) · Zbl 1178.90329
[41] Uciński, D., Optimal Measurement Methods for Distributed Parameter System Identification (2005), CRC Press: CRC Press Boca Raton · Zbl 1155.93003
[42] Uciński, D.; Bogacka, B., T-optimum designs for discriminating between two multiresponse dynamic models, J. R. Stat. Soc. Ser. B, 67, 3-18 (2005) · Zbl 1060.62084
[43] Ugray, Z.; Lasdon, L.; Plummer, J.; Glover, F.; Kelly, J.; Martí, R., A multistart scatter search heuristic for smooth NLP and MINLP problems, (Metaheuristic Optimization via Memory and Evolution (2005), Springer: Springer New York), 25-51 · Zbl 1072.90572
[44] Vajjah, P.; Duffull, S. B., A generalisation of T-optimality for discriminating between competing models with an application to pharmacokinetic studies, Pharmaceutical Stat., 11, 503-510 (2012)
[45] Waterhouse, T. H.; Eccleston, J. A.; Duffull, S. B., Optimal design criteria for discrimination and estimation in nonlinear models, J. Biopharm. Statist., 19, 386-402 (2009)
[46] Wiens, D. P., Robust discrimination designs, J. R. Stat. Soc. Ser. B, 71, 805-829 (2009) · Zbl 1248.62132
[47] Wynn, H. P., The sequential generation of D-optimum experimental designs, Ann. Math. Stat., 41, 1655-1664 (1970) · Zbl 0224.62038
[48] Žakovíc, S.; Rustem, B., Semi-Infinite Programming and applications to minimax problems, Ann. Oper. Res., 124, 81-110 (2003) · Zbl 1074.90554
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.