Geodesic
Preconditioners for least squares problems by LU factorization
Electronic transactions on numerical analysis, Tome 8 (1999), pp. 26-35
Iterative methods are often suitable for solving least-squares problems min , where kAx , bk A 2 2 m n is large and sparse. The use of the conjugate gradient method with a nonsingular square submatrix R A 2 1 n n of as preconditioner was first suggested by L$\ddot $auchli in 1961. This conjugate gradient method has recently R A been extended by Yuan to generalized least-squares problems.
Classification : 65F10, 65F20
Keywords: linear least squares, preconditioner, conjugate gradient method, LU factorization 
@article{ETNA_1999__8__a7,
     author = {Bj\"orck,  \r{A}ke and Yuan,  J.Y.},
     title = {Preconditioners for least squares problems by {LU} factorization},
     journal = {Electronic transactions on numerical analysis},
     pages = {26--35},
     year = {1999},
     volume = {8},
     zbl = {0924.65034},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/ETNA_1999__8__a7/}
}
TY  - JOUR
AU  - Björck,  Åke
AU  - Yuan,  J.Y.
TI  - Preconditioners for least squares problems by LU factorization
JO  - Electronic transactions on numerical analysis
PY  - 1999
SP  - 26
EP  - 35
VL  - 8
UR  - http://geodesic.mathdoc.fr/item/ETNA_1999__8__a7/
LA  - en
ID  - ETNA_1999__8__a7
ER  - 
%0 Journal Article
%A Björck,  Åke
%A Yuan,  J.Y.
%T Preconditioners for least squares problems by LU factorization
%J Electronic transactions on numerical analysis
%D 1999
%P 26-35
%V 8
%U http://geodesic.mathdoc.fr/item/ETNA_1999__8__a7/
%G en
%F ETNA_1999__8__a7
Björck,  Åke; Yuan,  J.Y. Preconditioners for least squares problems by LU factorization. Electronic transactions on numerical analysis, Tome 8 (1999), pp. 26-35. http://geodesic.mathdoc.fr/item/ETNA_1999__8__a7/