论文信息 - Bounding the growth factor in Gaussian elimination for Buckley's class of complex symmetric matrices

Bounding the growth factor in Gaussian elimination for Buckley's class of complex symmetric matrices

Let Mn(C) be the set of n × n complex matrices and A be a non-singular matrix in Mn(C). Consider the linear system Ax = b (1) and letA = ( a (k) ij ) be the matrix obtained fromA through thek steps of Gaussian elimination applied to A; in particular,A(n−1) is the upper triangular matrix resulting from the LU factorization ofA. The quantityρn(A) defined by ρn(A) = max i,j,k |a ij | max i,j |aij | (2)

Khakim D. Ikramov | Andrey B. Kucherov

[1] Nicholas J. Higham,et al. Factorizing complex symmetric matrices with positive definite real and imaginary parts , 1998, Math. Comput..

[2] A. Buckley,et al. On the Solution of Certain Skew Symmetric Linear Systems , 1977 .

[3] Roy Mathias,et al. Matrices with Positive Definite Hermitian Part: Inequalities and Linear Systems , 1992, SIAM J. Matrix Anal. Appl..

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