Minimizing the Condition Number to Construct Design Points for Polynomial Regression Models

In this paper we study a new optimality criterion, the $K$-optimality criterion, for constructing optimal experimental designs for polynomial regression models. We focus on the $p$th order polynomial regression model with symmetric design space $[-1,1]$. For this model, we show that there is always a symmetric $K$-optimal design with exactly $p+1$ support points including the boundary points $-1$ and $1$. It is well known that the condition number for a positive definite matrix as the ratio of the maximum eigenvalue to the minimum eigenvalue is usually nonsmooth. We show that for our model, the condition number of the information matrix is continuously differentiable. Theoretical $K$-optimal designs are derived for $p=1$ and $2$. Numerical results are presented for $3 \le p \le 10$.

[1]  W. J. Studden,et al.  Tchebycheff Systems: With Applications in Analysis and Statistics. , 1967 .

[2]  Michael L. Overton,et al.  Large-Scale Optimization of Eigenvalues , 1990, SIAM J. Optim..

[3]  Henry Wolkowicz,et al.  Handbook of Semidefinite Programming , 2000 .

[4]  Chen Greif,et al.  Minimizing the Condition Number for Small Rank Modifications , 2006, SIAM J. Matrix Anal. Appl..

[5]  M. Laurent Sums of Squares, Moment Matrices and Optimization Over Polynomials , 2009 .

[6]  Xiaojun Chen,et al.  Minimizing the Condition Number of a Gram Matrix , 2011, SIAM J. Optim..

[7]  J. Brian Gray,et al.  Introduction to Linear Regression Analysis , 2002, Technometrics.

[8]  J. Danskin The Theory of Max-Min and its Application to Weapons Allocation Problems , 1967 .

[9]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[10]  Steven E. Rigdon,et al.  Model-Oriented Design of Experiments , 1997, Technometrics.

[11]  R. Bass,et al.  Review: P. Billingsley, Convergence of probability measures , 1971 .

[12]  Allan Pinkus Totally Positive Matrices , 2009 .

[13]  Yu. S. Ledyaev,et al.  Nonsmooth analysis and control theory , 1998 .

[14]  E. Yaz Linear Matrix Inequalities In System And Control Theory , 1998, Proceedings of the IEEE.

[15]  F. Pukelsheim Optimal Design of Experiments , 1993 .

[16]  M. Morari,et al.  Minimizing the Euclidean Condition Number , 1994 .

[17]  Gene H. Golub,et al.  Matrix computations , 1983 .

[18]  W. J. Studden,et al.  Theory Of Optimal Experiments , 1972 .

[19]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[20]  C. Kaplan,et al.  Use of condition numbers for shortcut experimental design , 1987 .

[21]  Ting Kei Pong,et al.  Minimizing Condition Number via Convex Programming , 2011, SIAM J. Matrix Anal. Appl..

[22]  Jane J. Ye,et al.  Optimizing Condition Numbers , 2009, SIAM J. Optim..