NAG Library Routine Document
F08NBF (DGEEVX)
1 Purpose
F08NBF (DGEEVX) computes the eigenvalues and, optionally, the left and/or right eigenvectors for an by real nonsymmetric matrix .
Optionally, it also computes a balancing transformation to improve the conditioning of the eigenvalues and eigenvectors, reciprocal condition numbers for the eigenvalues, and reciprocal condition numbers for the right eigenvectors.
2 Specification
SUBROUTINE F08NBF ( |
BALANC, JOBVL, JOBVR, SENSE, N, A, LDA, WR, WI, VL, LDVL, VR, LDVR, ILO, IHI, SCALE, ABNRM, RCONDE, RCONDV, WORK, LWORK, IWORK, INFO) |
INTEGER |
N, LDA, LDVL, LDVR, ILO, IHI, LWORK, IWORK(*), INFO |
REAL (KIND=nag_wp) |
A(LDA,*), WR(*), WI(*), VL(LDVL,*), VR(LDVR,*), SCALE(*), ABNRM, RCONDE(*), RCONDV(*), WORK(max(1,LWORK)) |
CHARACTER(1) |
BALANC, JOBVL, JOBVR, SENSE |
|
The routine may be called by its
LAPACK
name dgeevx.
3 Description
The right eigenvector
of
satisfies
where
is the
th eigenvalue of
. The left eigenvector
of
satisfies
where
denotes the conjugate transpose of
.
Balancing a matrix means permuting the rows and columns to make it more nearly upper triangular, and applying a diagonal similarity transformation
, where
is a diagonal matrix, with the aim of making its rows and columns closer in norm and the condition numbers of its eigenvalues and eigenvectors smaller. The computed reciprocal condition numbers correspond to the balanced matrix. Permuting rows and columns will not change the condition numbers (in exact arithmetic) but diagonal scaling will. For further explanation of balancing, see Section 4.8.1.2 of
Anderson et al. (1999).
Following the optional balancing, the matrix is first reduced to upper Hessenberg form by means of unitary similarity transformations, and the algorithm is then used to further reduce the matrix to upper triangular Schur form, , from which the eigenvalues are computed. Optionally, the eigenvectors of are also computed and backtransformed to those of .
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
http://www.netlib.org/lapack/lug
Golub G H and Van Loan C F (1996) Matrix Computations (3rd Edition) Johns Hopkins University Press, Baltimore
5 Parameters
- 1: BALANC – CHARACTER(1)Input
On entry: indicates how the input matrix should be diagonally scaled and/or permuted to improve the conditioning of its eigenvalues.
- Do not diagonally scale or permute.
- Perform permutations to make the matrix more nearly upper triangular. Do not diagonally scale.
- Diagonally scale the matrix, i.e., replace by , where is a diagonal matrix chosen to make the rows and columns of more equal in norm. Do not permute.
- Both diagonally scale and permute .
Computed reciprocal condition numbers will be for the matrix after balancing and/or permuting. Permuting does not change condition numbers (in exact arithmetic), but balancing does.
Constraint:
, , or .
- 2: JOBVL – CHARACTER(1)Input
On entry: if
, the left eigenvectors of
are not computed.
If , the left eigenvectors of are computed.
If
or
,
JOBVL must be set to
.
Constraint:
or .
- 3: JOBVR – CHARACTER(1)Input
On entry: if
, the right eigenvectors of
are not computed.
If , the right eigenvectors of are computed.
If
or
,
JOBVR must be set to
.
Constraint:
or .
- 4: SENSE – CHARACTER(1)Input
On entry: determines which reciprocal condition numbers are computed.
- None are computed.
- Computed for eigenvalues only.
- Computed for right eigenvectors only.
- Computed for eigenvalues and right eigenvectors.
If or , both left and right eigenvectors must also be computed ( and ).
Constraint:
, , or .
- 5: N – INTEGERInput
On entry: , the order of the matrix .
Constraint:
.
- 6: A(LDA,) – REAL (KIND=nag_wp) arrayInput/Output
-
Note: the second dimension of the array
A
must be at least
.
On entry: the by matrix .
On exit:
A has been overwritten. If
or
,
contains the real Schur form of the balanced version of the input matrix
.
- 7: LDA – INTEGERInput
On entry: the first dimension of the array
A as declared in the (sub)program from which F08NBF (DGEEVX) is called.
Constraint:
.
- 8: WR() – REAL (KIND=nag_wp) arrayOutput
- 9: WI() – REAL (KIND=nag_wp) arrayOutput
-
Note: the dimension of the arrays
WR and
WI
must be at least
.
On exit:
WR and
WI contain the real and imaginary parts, respectively, of the computed eigenvalues. Complex conjugate pairs of eigenvalues appear consecutively with the eigenvalue having the positive imaginary part first.
- 10: VL(LDVL,) – REAL (KIND=nag_wp) arrayOutput
-
Note: the second dimension of the array
VL
must be at least
if
, and at least
otherwise.
On exit: if
, the left eigenvectors
are stored one after another in the columns of
VL, in the same order as their corresponding eigenvalues. If the
th eigenvalue is real, then
, the
th column of
VL. If the
th and
st eigenvalues form a complex conjugate pair, then
and
.
If
,
VL is not referenced.
- 11: LDVL – INTEGERInput
On entry: the first dimension of the array
VL as declared in the (sub)program from which F08NBF (DGEEVX) is called.
Constraints:
- if , ;
- otherwise .
- 12: VR(LDVR,) – REAL (KIND=nag_wp) arrayOutput
-
Note: the second dimension of the array
VR
must be at least
if
, and at least
otherwise.
On exit: if
, the right eigenvectors
are stored one after another in the columns of
VR, in the same order as their corresponding eigenvalues. If the
th eigenvalue is real, then
, the
th column of
VR. If the
th and
st eigenvalues form a complex conjugate pair, then
and
.
If
,
VR is not referenced.
- 13: LDVR – INTEGERInput
On entry: the first dimension of the array
VR as declared in the (sub)program from which F08NBF (DGEEVX) is called.
Constraints:
- if , ;
- otherwise .
- 14: ILO – INTEGEROutput
- 15: IHI – INTEGEROutput
On exit:
ILO and
IHI are integer values determined when
was balanced. The balanced
has
if
and
or
.
- 16: SCALE() – REAL (KIND=nag_wp) arrayOutput
-
Note: the dimension of the array
SCALE
must be at least
.
On exit: details of the permutations and scaling factors applied when balancing
.
If
is the index of the row and column interchanged with row and column
, and
is the scaling factor applied to row and column
, then
- , for ;
- , for ;
- , for .
The order in which the interchanges are made is
N to
, then
to
.
- 17: ABNRM – REAL (KIND=nag_wp)Output
On exit: the -norm of the balanced matrix (the maximum of the sum of absolute values of elements of any column).
- 18: RCONDE() – REAL (KIND=nag_wp) arrayOutput
-
Note: the dimension of the array
RCONDE
must be at least
.
On exit: is the reciprocal condition number of the th eigenvalue.
- 19: RCONDV() – REAL (KIND=nag_wp) arrayOutput
-
Note: the dimension of the array
RCONDV
must be at least
.
On exit: is the reciprocal condition number of the th right eigenvector.
- 20: WORK() – REAL (KIND=nag_wp) arrayWorkspace
On exit: if
,
contains the minimum value of
LWORK required for optimal performance.
- 21: LWORK – INTEGERInput
On entry: the dimension of the array
WORK as declared in the (sub)program from which F08NBF (DGEEVX) is called.
If
, 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 must generally be larger than the minimum, increase
LWORK by, say,
, where
is the optimal
block size for
F08NEF (DGEHRD).
Constraints:
- if and ,
- if , ;
- otherwise ;
- if or ,
- if or , ;
- otherwise .
- 22: IWORK() – INTEGER arrayWorkspace
-
Note: the dimension of the array
IWORK
must be at least
.
If
or
,
IWORK is not referenced.
- 23: INFO – INTEGEROutput
On exit:
unless the routine detects an error (see
Section 6).
6 Error Indicators and Warnings
Errors or warnings detected by the routine:
If , argument had an illegal value. An explanatory message is output, and execution of the program is terminated.
If
, the
algorithm failed to compute all the eigenvalues, and no eigenvectors or condition numbers have been computed; elements
and
of
WRWI
contain eigenvalues which have converged.
7 Accuracy
The computed eigenvalues and eigenvectors are exact for a nearby matrix
, where
and
is the
machine precision. See Section 4.8 of
Anderson et al. (1999) for further details.
Each eigenvector is normalized to have Euclidean norm equal to unity and the element of largest absolute value real and positive.
The total number of floating point operations is proportional to .
The complex analogue of this routine is
F08NPF (ZGEEVX).
9 Example
This example finds all the eigenvalues and right eigenvectors of the matrix
together with estimates of the condition number and forward error bounds for each eigenvalue and eigenvector. The option to balance the matrix is used. In order to compute the condition numbers of the eigenvalues, the left eigenvectors also have to be computed, but they are not printed out in this example.
Note that the block size (NB) of assumed in this example is not realistic for such a small problem, but should be suitable for large problems.
9.1 Program Text
Program Text (f08nbfe.f90)
9.2 Program Data
Program Data (f08nbfe.d)
9.3 Program Results
Program Results (f08nbfe.r)