NAG FL Interface
f08anf (zgels)

1 Purpose

f08anf solves linear least squares problems of the form
minx b-Ax2   or   minx b-AHx2 ,  
where A is an m by n complex matrix of full rank, using a QR or LQ factorization of A.

2 Specification

Fortran Interface
Subroutine f08anf ( trans, m, n, nrhs, a, lda, b, ldb, work, lwork, info)
Integer, Intent (In) :: m, n, nrhs, lda, ldb, lwork
Integer, Intent (Out) :: info
Complex (Kind=nag_wp), Intent (Inout) :: a(lda,*), b(ldb,*)
Complex (Kind=nag_wp), Intent (Out) :: work(max(1,lwork))
Character (1), Intent (In) :: trans
C Header Interface
#include <nag.h>
void  f08anf_ (const char *trans, const Integer *m, const Integer *n, const Integer *nrhs, Complex a[], const Integer *lda, Complex b[], const Integer *ldb, Complex work[], const Integer *lwork, Integer *info, const Charlen length_trans)
The routine may be called by the names f08anf, nagf_lapackeig_zgels or its LAPACK name zgels.

3 Description

The following options are provided:
  1. 1.If trans='N' and mn: find the least squares solution of an overdetermined system, i.e., solve the least squares problem
    minx b-Ax2 .  
  2. 2.If trans='N' and m<n: find the minimum norm solution of an underdetermined system Ax=b.
  3. 3.If trans='C' and mn: find the minimum norm solution of an undetermined system AHx=b.
  4. 4.If trans='C' and m<n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem
    minx b-AHx2 .  
Several right-hand side vectors b and solution vectors x can be handled in a single call; they are stored as the columns of the m by r right-hand side matrix B and the n by r solution matrix X.

4 References

Anderson E, Bai Z, Bischof C, Blackford S, Demmel J, Dongarra J J, Du Croz J J, Greenbaum A, Hammarling S, McKenney A and Sorensen D (1999) LAPACK Users' Guide (3rd Edition) SIAM, Philadelphia https://www.netlib.org/lapack/lug
Golub G H and Van Loan C F (1996) Matrix Computations (3rd Edition) Johns Hopkins University Press, Baltimore

5 Arguments

1: trans Character(1) Input
On entry: if trans='N', the linear system involves A.
If trans='C', the linear system involves AH.
Constraint: trans='N' or 'C'.
2: m Integer Input
On entry: m, the number of rows of the matrix A.
Constraint: m0.
3: n Integer Input
On entry: n, the number of columns of the matrix A.
Constraint: n0.
4: nrhs Integer Input
On entry: r, the number of right-hand sides, i.e., the number of columns of the matrices B and X.
Constraint: nrhs0.
5: alda* Complex (Kind=nag_wp) array Input/Output
Note: the second dimension of the array a must be at least max1,n.
On entry: the m by n matrix A.
On exit: if mn, a is overwritten by details of its QR factorization, as returned by f08asf.
If m<n, a is overwritten by details of its LQ factorization, as returned by f08avf.
6: lda Integer Input
On entry: the first dimension of the array a as declared in the (sub)program from which f08anf is called.
Constraint: ldamax1,m.
7: bldb* Complex (Kind=nag_wp) array Input/Output
Note: the second dimension of the array b must be at least max1,nrhs.
On entry: the matrix B of right-hand side vectors, stored in columns; b is m by r if trans='N', or n by r if trans='C'.
On exit: b is overwritten by the solution vectors, x, stored in columns:
  • if trans='N' and mn, or trans='C' and m<n, elements 1 to minm,n in each column of b contain the least squares solution vectors; the residual sum of squares for the solution is given by the sum of squares of the modulus of elements minm,n+1 to maxm,n in that column;
  • otherwise, elements 1 to maxm,n in each column of b contain the minimum norm solution vectors.
8: ldb Integer Input
On entry: the first dimension of the array b as declared in the (sub)program from which f08anf is called.
Constraint: ldbmax1,m,n.
9: workmax1,lwork Complex (Kind=nag_wp) array Workspace
On exit: if info=0, the real part of work1 contains the minimum value of lwork required for optimal performance.
10: lwork Integer Input
On entry: the dimension of the array work as declared in the (sub)program from which f08anf is called.
If lwork=-1, a workspace query is assumed; the routine only calculates the optimal size of the work array, returns this value as the first entry of the work array, and no error message related to lwork is issued.
Suggested value: for optimal performance, lwork minm,n + max1,m,n,nrhs × nb , where nb is the optimal block size.
Constraint: lworkminm,n+max1,m,n,nrhs or lwork=-1.
11: info Integer Output
On exit: info=0 unless the routine detects an error (see Section 6).

6 Error Indicators and Warnings

info<0
If info=-i, argument i had an illegal value. An explanatory message is output, and execution of the program is terminated.
info>0
Diagonal element value of the triangular factor of A is zero, so that A does not have full rank; the least squares solution could not be computed.

7 Accuracy

See Section 4.5 of Anderson et al. (1999) for details of error bounds.

8 Parallelism and Performance

f08anf is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.
f08anf makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the X06 Chapter Introduction for information on how to control and interrogate the OpenMP environment used within this routine. Please also consult the Users' Note for your implementation for any additional implementation-specific information.

9 Further Comments

The total number of floating-point operations required to factorize A is approximately 83 n2 3m-n if mn and 83 m2 3n-m otherwise. Following the factorization the solution for a single vector x requires O minm2,n2 operations.
The real analogue of this routine is f08aaf.

10 Example

This example solves the linear least squares problem
minx b-Ax2 ,  
where
A = 0.96-0.81i -0.03+0.96i -0.91+2.06i -0.05+0.41i -0.98+1.98i -1.20+0.19i -0.66+0.42i -0.81+0.56i 0.62-0.46i 1.01+0.02i 0.63-0.17i -1.11+0.60i -0.37+0.38i 0.19-0.54i -0.98-0.36i 0.22-0.20i 0.83+0.51i 0.20+0.01i -0.17-0.46i 1.47+1.59i 1.08-0.28i 0.20-0.12i -0.07+1.23i 0.26+0.26i  
and
b = -2.09+1.93i 3.34-3.53i -4.94-2.04i 0.17+4.23i -5.19+3.63i 0.98+2.53i .  
The square root of the residual sum of squares is also output.
Note that the block size (NB) of 64 assumed in this example is not realistic for such a small problem, but should be suitable for large problems.

10.1 Program Text

Program Text (f08anfe.f90)

10.2 Program Data

Program Data (f08anfe.d)

10.3 Program Results

Program Results (f08anfe.r)