Electronic Journal of Statistics

Optimal experimental design that minimizes the width of simultaneous confidence bands

Satoshi Kuriki and Henry P. Wynn

Full-text: Open access


We propose an optimal experimental design for a curvilinear regression model that minimizes the band-width of simultaneous confidence bands. Simultaneous confidence bands for curvilinear regression are constructed by evaluating the volume of a tube about a curve that is defined as a trajectory of a regression basis vector (Naiman, 1986). The proposed criterion is constructed based on the volume of a tube, and the corresponding optimal design that minimizes the volume of tube is referred to as the tube-volume optimal (TV-optimal) design. For Fourier and weighted polynomial regressions, the problem is formalized as one of minimization over the cone of Hankel positive definite matrices, and the criterion to minimize is expressed as an elliptic integral. We show that the Möbius group keeps our problem invariant, and hence, minimization can be conducted over cross-sections of orbits. We demonstrate that for the weighted polynomial regression and the Fourier regression with three bases, the tube-volume optimal design forms an orbit of the Möbius group containing D-optimal designs as representative elements.

Article information

Electron. J. Statist., Volume 13, Number 1 (2019), 1099-1134.

Received: October 2018
First available in Project Euclid: 5 April 2019

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62K05: Optimal designs

D-optimality Fourier regression Hankel matrix Möbius group volume-of-tube method weighted polynomial regression

Creative Commons Attribution 4.0 International License.


Kuriki, Satoshi; Wynn, Henry P. Optimal experimental design that minimizes the width of simultaneous confidence bands. Electron. J. Statist. 13 (2019), no. 1, 1099--1134. doi:10.1214/19-EJS1546. https://projecteuclid.org/euclid.ejs/1554429625

Export citation


  • [1] Adler, R. J. and Taylor, J. E. (2007)., Random Fields and Geometry, Springer.
  • [2] Dette, H., Haines, L. M. and Imhof, L. (1999). Optimal designs for rational models and weighted polynomial regression, The Annals of Statistics, 27 (4), 1272–1293.
  • [3] Dette, H. and Melas, V. B. (2003). Optimal designs for estimating individual coefficients in Fourier regression models, The Annals of Statistics, 31 (5), 1669–1692.
  • [4] Dette, H. and Schorning, K. (2016). Optimal designs for comparing curves, The Annals of Statistics, 44 (3), 1103–1130.
  • [5] Dette, H., Schorning, K. and Konstantinou, M. (2017). Optimal designs for comparing regression models with correlated observations, Computational Statistics & Data Analysis, 113, 273–286.
  • [6] Drton, M., Sturmfels, B. and Sullivant, S. (2009)., Lectures on Algebraic Statistics, Oberwolfach Seminars 39, Springer.
  • [7] van Dyk, D. A. (2014). The role of statistics in the discovery of a Higgs boson, Annual Review of Statistics and Its Application, 1, 41–59.
  • [8] Fang, K.-T. and Zhang, Y.-T. (1990)., Generalized Multivariate Analysis, Science Press and Springer.
  • [9] Gould, H. W. (2010)., Combinatorial Identities: Table I: Intermediate Techniques for Summing Finite Series: From the seven unpublished manuscripts of H. W. Gould Edited and Compiled by Jocelyn Quaintance. http://www.math.wvu.edu/~gould/Vol.4.PDF
  • [10] Gross, K. I. and Holman III, W. J. (1980). Matrix-valued special functions and representation theory of the conformal group, I: The generalized gamma function, Transactions of the American Mathematical Society, 258 (2), 319–350.
  • [11] Guest, P. G. (1958). The spacing of observations in polynomial regression, The Annals of Mathematical Statistics, 29 (1), 294–299.
  • [12] Heinig, G. and Rost, K. (1989). Matrices with displacement structure, generalized Bezoutians, and Möbius transformations, Operator Theory: Advances and Applications, 40, 203–230.
  • [13] Heinig, G. and Rost, K. (2010). Introduction to Bezoutians, Operator Theory: Advances and Applications, 199, 25–118.
  • [14] Johansen, S. and Johnstone, I. M. (1990). Hotelling’s theorem on the volume of tubes: Some illustrations in simultaneous inference and data analysis, The Annals of Statistics, 18 (2), 652–684.
  • [15] Johnstone, I. and Siegmund, D. (1989). On Hotelling’s formula for the volume of tubes and Naiman’s inequality, The Annals of Statistics, 17 (1), 184–194.
  • [16] Karlin, S. and Studden, W. (1966)., Tchebycheff Systems: With Applications in Analysis and Statistics, Interscience Publishers, Wiley.
  • [17] Kato, N. and Kuriki, S. (2013). Likelihood ratio tests for positivity in polynomial regressions, Journal of Multivariate Analysis, 115, 334–346.
  • [18] Kato, S. and McCullagh, P. (2014). A characterization of a Cauchy family on the complex space, arXiv:1402.1905 [math.ST].
  • [19] Kawakubo, K. (1992)., The Theory of Transformation Groups, Oxford University Press.
  • [20] Kiefer, J. and Wolfowitz, J. (1959). Optimum designs in regression problems, The Annals of Mathematical Statistics, 30 (2), 271–294.
  • [21] Kuriki, S. and Takemura, A. (2001). Tail probabilities of the maxima of multilinear forms and their applications, The Annals of Statistics, 29 (2), 328–371.
  • [22] Kuriki, S. and Takemura, A. (2009). Volume of tubes and the distribution of the maximum of a Gaussian random field, Selected Papers on Probability and Statistics, American Mathematical Society Translations Series 2, Vol. 227, No. 2, 25–48.
  • [23] Lasserre, J. B. (2009)., Moments, Positive Polynomials and Their Applications, Imperial College Press.
  • [24] Liu, W. (2010)., Simultaneous Inference in Regression, Chapman & Hall/CRC.
  • [25] Lu, X. and Kuriki, S. (2017). Simultaneous confidence bands for contrasts between several nonlinear regression curves, Journal of Multivariate Analysis, 155, 83–104.
  • [26] Mackey, D. S., Mackey, N., Mehl, C. and Mehrmann, V. (2015). Möbius transformations of matrix polynomials, Linear Algebra and Its Applications, 470, 120–184.
  • [27] McCullagh, P. (1996). Möbius transformation and Cauchy parameter estimation, The Annals of Statististics, 24 (2), 787–808.
  • [28] Moriguti, S., Udagawa, K. and Hitotumatu, S. (1957)., Iwanami Sugaku Koshiki, II (in Japanese), Iwanami Shoten.
  • [29] Naiman, D. Q. (1986). Conservative confidence bands in curvilinear regression, The Annals of Statistics, 14 (3), 896–906.
  • [30] Prasolov, V. V. (2004)., Polynomials, Algorithms and Computation in Mathematics 11, Springer.
  • [31] Pukelsheim, F. (2006)., Optimal Design of Experiments, Classics in Applied Mathematics 50, SIAM.
  • [32] Scheffé, H. (1959)., The Analysis of Variance, Wiley.
  • [33] Sun, J. (1993). Tail probabilities of the maxima of Gaussian random fields, The Annals of Probability, 21 (1), 34–71.
  • [34] Sun, J. and Loader, C. R. (1994). Simultaneous confidence bands for linear regression and smoothing, The Annals of Probability, 22 (3), 1328–1345.
  • [35] Takemura, A. and Kuriki, S. (2002). On the equivalence of the tube and Euler characteristic methods for the distribution of the maximum of Gaussian fields over piecewise smooth domains, The Annals of Applied Probability, 12 (2), 768–796.
  • [36] Working, H. and Hotelling, H. (1929). Applications of the theory of error to the interpretation of trends, Journal of the American Statistical Association, 24 (165), 73–85.
  • [37] Wynn, H. P. (1985). Jack Kiefer’s contributions to experimental design, in, Jack Carl Kiefer Collected Papers III: Design of Experiments, L. D. Brown, O. Olkin, J. Sacks, H, P. Wynn eds, Springer and the Institute of Mathematical Statistics, 1985, pages xvii–xxiii.
  • [38] Wynn, H. P. and Bloomfield, P. (1971). Simultaneous confidence bands in regression analysis (with discussions), Journal of the Royal Statistical Society, Ser. B, 33 (2), 202–221.