TOPICS
Search

LQ Decomposition


An LQ decomposition of a real m×n matrix A with m<=n is a matrix decomposition

 A=LQ,

where L is an m×m lower triangular matrix and the m×n matrix Q has rows forming an orthonormal set, so that QQ^T=I_m. This is the reduced form. In the full form, the factor Q is an n×n orthogonal matrix and the left factor is (L 0). For complex matrices, conjugate transpose replaces transpose, and the full factor is a unitary matrix.

An LQ decomposition follows by transposing a QR decomposition of A^T. If m<n and A has matrix rank m, it gives the solution of Ax=b having the smallest Euclidean norm,

 x=Q^TL^(-1)b,

using the reduced factors. In computation, the linear system of equations with matrix L is solved without forming its matrix inverse.


See also

Linear System of Equations, Lower Triangular Matrix, Matrix Decomposition, QR Decomposition

Explore with Wolfram|Alpha

References

Anderson, E.; Bai, Z.; Bischof, C.; Blackford, S.; Demmel, J.; Dongarra, J.; Du Croz, J.; Greenbaum, A.; Hammarling, S.; McKenney, A.; and Sorensen, D. "LQ Factorization." In LAPACK Users' Guide, 3rd ed. Philadelphia, PA: SIAM, 1999. https://www.netlib.org/lapack/lug/node41.html.

Cite this as:

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

Subject classifications