NAG Library Routine Document
F07PUF (ZHPCON)
1 Purpose
F07PUF (ZHPCON) estimates the condition number of a complex Hermitian indefinite matrix
, where
has been factorized by
F07PRF (ZHPTRF), using packed storage.
2 Specification
INTEGER |
N, IPIV(*), INFO |
REAL (KIND=nag_wp) |
ANORM, RCOND |
COMPLEX (KIND=nag_wp) |
AP(*), WORK(2*N) |
CHARACTER(1) |
UPLO |
|
The routine may be called by its
LAPACK
name zhpcon.
3 Description
F07PUF (ZHPCON) estimates the condition number (in the
-norm) of a complex Hermitian indefinite matrix
:
Since
is Hermitian,
.
Because is infinite if is singular, the routine actually returns an estimate of the reciprocal of .
The routine should be preceded by a call to
F06UDF to compute
and a call to
F07PRF (ZHPTRF) to compute the Bunch–Kaufman factorization of
. The routine then uses Higham's implementation of Hager's method (see
Higham (1988)) to estimate
.
4 References
Higham N J (1988) FORTRAN codes for estimating the one-norm of a real or complex matrix, with applications to condition estimation ACM Trans. Math. Software 14 381–396
5 Parameters
- 1: – CHARACTER(1)Input
-
On entry: specifies how
has been factorized.
- , where is upper triangular.
- , where is lower triangular.
Constraint:
or .
- 2: – INTEGERInput
-
On entry: , the order of the matrix .
Constraint:
.
- 3: – COMPLEX (KIND=nag_wp) arrayInput
-
Note: the dimension of the array
AP
must be at least
.
On entry: the factorization of
stored in packed form, as returned by
F07PRF (ZHPTRF).
- 4: – INTEGER arrayInput
-
Note: the dimension of the array
IPIV
must be at least
.
On entry: details of the interchanges and the block structure of
, as returned by
F07PRF (ZHPTRF).
- 5: – REAL (KIND=nag_wp)Input
-
On entry: the
-norm of the
original matrix
, which may be computed by calling
F06UDF with its parameter
.
ANORM must be computed either
before calling
F07PRF (ZHPTRF) or else from a
copy of the original matrix
.
Constraint:
.
- 6: – REAL (KIND=nag_wp)Output
-
On exit: an estimate of the reciprocal of the condition number of
.
RCOND is set to zero if exact singularity is detected or the estimate underflows. If
RCOND is less than
machine precision,
is singular to working precision.
- 7: – COMPLEX (KIND=nag_wp) arrayWorkspace
-
- 8: – INTEGEROutput
On exit:
unless the routine detects an error (see
Section 6).
6 Error Indicators and Warnings
If , argument had an illegal value. An explanatory message is output, and execution of the program is terminated.
7 Accuracy
The computed estimate
RCOND is never less than the true value
, and in practice is nearly always less than
, although examples can be constructed where
RCOND is much larger.
8 Parallelism and Performance
F07PUF (ZHPCON) is not threaded by NAG in any implementation.
F07PUF (ZHPCON) 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.
A call to F07PUF (ZHPCON) involves solving a number of systems of linear equations of the form
; the number is usually
and never more than
. Each solution involves approximately
real floating-point operations but takes considerably longer than a call to
F07PSF (ZHPTRS) with one right-hand side, because extra care is taken to avoid overflow when
is approximately singular.
The real analogue of this routine is
F07PGF (DSPCON).
10 Example
This example estimates the condition number in the
-norm (or
-norm) of the matrix
, where
Here
is Hermitian indefinite, stored in packed form, and must first be factorized by
F07PRF (ZHPTRF). The true condition number in the
-norm is
.
10.1 Program Text
Program Text (f07pufe.f90)
10.2 Program Data
Program Data (f07pufe.d)
10.3 Program Results
Program Results (f07pufe.r)