TOPICS
Search

QZ Decomposition


The QZ decomposition, also called the generalized Schur decomposition, of two square complex matrices A and B consists of unitary matrices Q and Z such that

Q^*AZ=S
(1)
Q^*BZ=T,
(2)

where S and T are upper triangular matrices. If the matrix pencil A-lambdaB is regular, meaning that its determinant is not identically zero, its generalized eigenvalues are represented by the diagonal pairs (s_(ii),t_(ii)). A pair with t_(ii)!=0 gives lambda=s_(ii)/t_(ii), while t_(ii)=0 gives an infinite generalized eigenvalue. A singular matrix pencil can produce an indeterminate pair (0,0).

For real matrices, a real QZ decomposition uses orthogonal matrices and an upper quasitriangular S with an upper triangular matrix T. The diagonal-pair rule applies to the 1-by-1 blocks. The 2-by-2 blocks encode conjugate pairs of nonreal generalized eigenvalues, which cannot be read from individual diagonal ratios.


See also

Generalized Eigenvalue, Matrix Decomposition, Schur Decomposition

Explore with Wolfram|Alpha

References

Golub, G. H. and Van Loan, C. F. Matrix Computations, 4th ed. Baltimore, MD: Johns Hopkins University Press, 2013.

Cite this as:

Weisstein, Eric W. "QZ Decomposition." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/QZDecomposition.html

Subject classifications