Abstract
We consider T-optimal experiment design problems for discriminating multi-factor polynomial regression models where the design space is defined by polynomial inequalities and the regression parameters are constrained to given convex sets. Our proposed optimality criterion is formulated as a convex optimization problem with a moment cone constraint. When the regression models have one factor, an exact semidefinite representation of the moment cone constraint can be applied to obtain an equivalent semidefinite program. When there are two or more factors in the models, we apply a moment relaxation technique and approximate the moment cone constraint by a hierarchy of semidefinite-representable outer approximations. When the relaxation hierarchy converges, an optimal discrimination design can be recovered from the optimal moment matrix, and its optimality can be additionally confirmed by an equivalence theorem. The methodology is illustrated with several examples.
Similar content being viewed by others
References
ApS, M: The MOSEK optimization toolbox for MATLAB manual. Version 8.1. http://docs.mosek.com/8.1/toolbox/index.html. (2017)
Atkinson, A.C.: Optimum experimental designs for choosing between competitive and non competitive models of enzyme inhibition. Commun. Stat. Theory Methods 41(13–14), 2283–2296 (2012)
Atkinson, A., Cox, D.R.: Planning experiments for discriminating between models. J. R. Stat. Soc. Ser. B (Methodol.) 36(3), 321–348 (1974)
Atkinson, A.C., Fedorov, V.: The design of experiments for discriminating between two rival models. Biometrika 62(1), 57–70 (1975a)
Atkinson, A.C., Fedorov, V.V.: The design of experiments for discriminating between several rival models. Biometrika 62(2), 289–303 (1975b)
Bisschop, J: AIMMS optimization modeling. Lulu.com (2006)
Boyd, S., Vandenberghe, L.: Convex Optimization. Cambridge University Press, Cambridge (2004)
Carlos Monteiro Ponce de Leon, A.: Optimum experimental design for model discrimination and generalized linear models. Ph.D. thesis, London School of Economics and Political Science (United Kingdom) (1993)
De Castro, Y., Gamboa, F., Henrion, D., Hess, R., Lasserre, J.B.: D-optimal design for multivariate polynomial regression via the christoffel function and semidefinite relaxations. (2017) arXiv preprint arXiv:170301777
De Leon, A.P., Atkinson, A.C.: Optimum experimental design for discriminating between two rival models in the presence of prior information. Biometrika 78(3), 601–608 (1991)
Dette, H.: Discrimination designs for polynomial regression on compact intervals. Ann. Stat. 22(2), 890–903 (1994)
Dette, H., Melas, V.B.: Optimal designs for estimating individual coefficients in fourier regression models. Ann. Stat. 31(5), 1669–1692 (2003)
Dette, H., Titoff, S.: Optimal discrimination designs. Ann. Stat. 37(4), 2056–2082 (2009)
Dette, H., Melas, V.B., Shpilev, P.: T-optimal designs for discrimination between two polynomial models. Ann. Stat. 40(1), 188–205 (2012)
Dette, H., Guchenko, R., Melas, V., Wong, W.K.: Optimal discrimination designs for semi-parametric models. Biometrika 105(1), 185–197 (2018)
Diamond, S., Boyd, S.: Cvxpy: a python-embedded modeling language for convex optimization. J. Mach. Learn. Res. 17(1), 2909–2913 (2016)
Duarte, B.P., Wong, W.K., Atkinson, A.C.: A semi-infinite programming based algorithm for determining t-optimum designs for model discrimination. J. Multivar. Anal. 135, 11–24 (2015)
Duarte, B.P., Wong, W.K., Dette, H.: Adaptive grid semidefinite programming for finding optimal designs. Stat. Comput. 28(2), 441–460 (2018)
Fedorov, V.V.: The design of experiments in the multiresponse case. Theory Probab. Appl. 16(2), 323–332 (1971)
Fedorov, V.: Theory of Optimal Experiments. Elsevier, New York (1972)
Fedorov, V.V., Hackl, P.: Model-Oriented Design of Experiments, vol. 125. Springer, Berlin (2012)
Fedorov, V.V., Leonov, S.L.: Optimal Design for Nonlinear Response Models. CRC Press, Boca Raton (2013)
Fedorov, V.V., Malyutov, M.B.: Optimal designs in regression problems. Math Operationsforsch Statist 3(4), 281–308 (1972)
Gahinet, P., Nemirovskii, A., Laub, A.J., Chilali, M.: The lmi control toolbox. Decision and Control, 1994. In: Proceedings of the 33rd IEEE Conference on Decision and Control, vol. 3, pp. 2038–2041 (1994)
Goh, J., Sim, M.: Robust optimization made easy with rome. Oper. Res. 59(4), 973–985 (2011)
Grant, M., Boyd, S., Ye, Y.: Cvx: Matlab software for disciplined convex programming (2008)
Henrion, D., Lasserre, J.B.: Detecting global optimality and extracting solutions in gloptipoly 312, 293–310 (2005)
Henrion, D., Lasserre, J.B., Löfberg, J.: Gloptipoly 3: moments, optimization and semidefinite programming. Optim. Methods Softw. 24(4–5), 761–779 (2009)
Hess, R: Some approximation schemes in polynomial optimization. Ph.D. thesis, Université de Toulouse, Université Toulouse III-Paul Sabatier (2017)
Karlin, S., Studden, W.: Tchebycheff systems: with applications in analysis and statistics, Interscience, New York, vol. 15. Interscience Publishers (1966)
Lasserre, J.B.: Global optimization with polynomials and the problem of moments. SIAM J. Optim. 11(3), 796–817 (2001)
Lasserre, J.B.: Moments, Positive Polynomials and Their Applications, vol. 1. World Scientific, Singapore (2009)
Lasserre, J.B.: An Introduction to Polynomial and Semi-algebraic Optimization, vol. 52. Cambridge University Press, Cambridge (2015)
Lofberg, J: Yalmip: A toolbox for modeling and optimization in matlab. In: 2004 IEEE International Conference on Robotics and Automation, pp. 284–289 (2004)
López-Fidalgo, J., Tommasi, C., Trandafir, P.: An optimal experimental design criterion for discriminating between non-normal models. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 69(2), 231–242 (2007)
Nie, J.: The \({{\cal{A}}}\)-truncated k-moment problem. Found. Comput. Math. 14(6), 1243–1276 (2014)
Rockafellar, R.: Convex Analysis. Princeton University Press, Princeton (1970)
Scheiderer, C.: Spectrahedral shadows. SIAM J. Appl. Algebra Geom. 2(1), 26–44 (2018)
Shohat, J.A., Tamarkin, J.D.: The Problem of Moments, vol. 1. American Mathematical Society, Providence (1943)
Sturm, J.F.: Using sedumi 1.02, a matlab toolbox for optimization over symmetric cones. Optim. Methods Softw. 11(1–4), 625–653 (1999)
Toh, K.C., Todd, M.J., Tütüncü, R.H.: Sdpt3—a matlab software package for semidefinite programming, version 1.3. Optim. Methods Softw. 11(1–4), 545–581 (1999)
Uciński, D.: Optimal Measurement Methods for Distributed Parameter System Identification. CRC Press, Boca Raton (2004)
Uciński, D., Bogacka, B.: T-optimum designs for discrimination between two multiresponse dynamic models. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 67(1), 3–18 (2005)
Waterhouse, T., Eccleston, J., Duffull, S.: Optimal design criteria for discrimination and estimation in nonlinear models. J. Biopharm. Stat. 19(2), 386–402 (2009)
Wiens, D.P.: Robust discrimination designs. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 71(4), 805–829 (2009)
Wong, W.K., Chen, R.B., Huang, C.C., Wang, W.: A modified particle swarm optimization technique for finding optimal designs for mixture models. PLoS ONE 10(6), e0124720 (2015)
Wynn, H.P.: The sequential generation of d-optimum experimental designs. Ann. Math. Stat. 41(5), 1655–1664 (1970)
Yang, M., Biedermann, S., Tang, E.: On optimal designs for nonlinear models: a general and efficient algorithm. J. Am. Stat. Assoc. 108(504), 1411–1420 (2013)
Acknowledgements
All authors gratefully acknowledge partial support from a grant from the National Institute of General Medical Sciences of the National Institutes of Health under Award Number R01GM107639. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Dr. Lieven Vandenberghe is also partially by a National Science Foundation Grant 1509789. We wish to thank Dr. Didier Henrion for helpful discussions and advice on using the software package GloptiPoly3. We also thank Dr. Fabrice Gamboa for his helpful comments on an earlier version of this manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
We provide here an illustrative MATLAB code for finding the moment matrix of optimal discrimination design in Example 1.
Rights and permissions
About this article
Cite this article
Yue, Y., Vandenberghe, L. & Wong, W.K. T-optimal designs for multi-factor polynomial regression models via a semidefinite relaxation method. Stat Comput 29, 725–738 (2019). https://doi.org/10.1007/s11222-018-9834-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-018-9834-2