Association Journal of CSIAM
Supervised by Ministry of Education of PRC
Sponsored by Xi'an Jiaotong University
ISSN 1005-3085  CN 61-1269/O1

Chinese Journal of Engineering Mathematics ›› 2023, Vol. 40 ›› Issue (1): 123-134.doi: 10.3969/j.issn.1005-3085.2023.01.009

Previous Articles     Next Articles

Applications and Properties of Permutation Matrix for Mixture Experimental Designs

LI Junpeng1,2,   LI Guanghui2,   ZHANG Chongqi1   

  1. 1. School of Economics and Statistics, Guangzhou University, Guangzhou 510006;
    2. School of Science, Kaili University, Kaili 556011
  • Online:2023-02-15 Published:2023-04-11
  • Contact: C. Zhang. E-mail address: cqzhang@gzhu.edu.cn
  • Supported by:
    The National Natural Science Foundation of China (11901260; 12071096); the Foundation of Guizhou Science and Technology Department ([2020]1Y010).

Abstract:

In order to solve the problem of the complex form and complicated calculation of the design matrix for high-dimensional mixture experiments, the definition of the permutation matrix is proposed by using high-order mixture lattice points as the support set. Firstly, the method to construct the design matrix from the permutation matrix is discussed, and the representation of the information matrix and the correlation matrix are derived. Then, the properties of the determinant, trace, and eigenvalue of the permutation matrix are studied. Finally, the effectiveness of the permutation matrix for the analysis of mixture design is verified, and the necessity and feasibility of using the permutation matrix to construct a high-order lattice point set and an optimal design are further analyzed.

Key words: permutation matrix, information matrix, mixture experimental design, $D$-optimal design

CLC Number: