Infinity norm bounds for the inverse for GSDD1 matrices using scaling matrices

被引:0
|
作者
Dai, Ping-Fan [1 ,2 ]
Li, Jinping [3 ]
Zhao, Shaoyu [2 ]
机构
[1] Hanshan Normal Univ, Sch Math & Stat, Chaozhou 521041, Guangdong, Peoples R China
[2] Sanming Univ, Sch Informat Engn, Sanming 365004, Fujian, Peoples R China
[3] Hainan Univ, Sch Sci, Haikou 570228, Hainan, Peoples R China
关键词
Infinity norm bounds; SDD matrices; Error bounds; Linear complementarity problem; LINEAR COMPLEMENTARITY-PROBLEMS; ERROR-BOUNDS;
D O I
10.1007/s40314-022-02165-x
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Many subclasses of H -matrices have already been investigated in application areas of linear algebra. One of them is SDD1 matrices given in Pella (Adv Comput Math 35:357-373, 2011). In this paper, a new subclass of H -matrices, i.e., generalized SDD1(GSDD(1)) matrices, is considered. The relationship between GSDD(1) matrices and other subclasses of H -matrices is analyzed. Infinity norm bounds for the inverse of a GSDD(1) matrix A are given, using a scaling matrix that transforms A into a strictly diagonally dominant matrix. The given scaling matrix is also utilized to obtain error bounds for the linear complementarity problems when the related matrices are GSDD(1) matrices. Numerical examples show that the obtained results can improve other existing bounds.
引用
收藏
页数:21
相关论文
共 25 条