Alle boeken

Professioneel

Apps Over Coach Inloggen Begin met lezen

Quantitative Finance · Begrippenlijst

Wat is LU, Cholesky, QR and singular value decompositions?

Ook bekend als: LU factorisation · Cholesky factorisation · QR factorisation · singular value decomposition · tridiagonal matrix algorithm

Definition 25.7 Quantitative Methods · Hoofdstuk 25 — Floating Point and Numerical Linear Algebra

The LU factorisation writes PA=LUPA = LU with a permutation PP (partial pivoting), unit lower-triangular LL and upper-triangular UU. The Cholesky factorisation of a symmetric positive-definite matrix is A=LL⊤A = LL^\top with LL lower triangular. The QR factorisation writes A=QRA = QR with orthonormal columns in QQ and RR upper triangular. The singular value decomposition writes A=UΣV⊤A = U\Sigma V^\top with orthonormal UU, VV and a nonnegative diagonal Σ\Sigma. The tridiagonal matrix algorithm (Thomas’s algorithm) solves a tridiagonal system by one forward elimination and one back substitution, in O(n)O(n) operations.

Lees in het hoofdstuk →