void	f08ykc (Nag_OrderType order, Nag_SideType side, Nag_HowManyType how_many, const Nag_Boolean select[], Integer n, const double a[], Integer pda, const double b[], Integer pdb, double vl[], Integer pdvl, double vr[], Integer pdvr, Integer mm, Integer m, NagError fail)

The function may be called by the names: f08ykc, nag_lapackeig_dtgevc or nag_dtgevc.

3 Description

f08ykc computes some or all of the right and/or left generalized eigenvectors of the matrix pair

(A, B)

which is assumed to be in generalized upper Schur form. If the matrix pair

(A, B)

is not in the generalized upper Schur form, then f08xec should be called before invoking f08ykc.

The right generalized eigenvector

x

and the left generalized eigenvector

y

(A, B)

corresponding to a generalized eigenvalue

λ

are defined by

(A - λ B) x = 0

and

y^{H} (A - λ B) = 0 .

If a generalized eigenvalue is determined as

0 / 0

, which is due to zero diagonal elements at the same locations in both

A

and

B

, a unit vector is returned as the corresponding eigenvector.

Note that the generalized eigenvalues are computed using f08xec but f08ykc does not explicitly require the generalized eigenvalues to compute eigenvectors. The ordering of the eigenvectors is based on the ordering of the eigenvalues as computed by f08ykc.

If all eigenvectors are requested, the function may either return the matrices

X

and/or

Y

of right or left eigenvectors of

(A, B)

, or the products

Z X

and/or

Q Y

, where

Z

and

Q

are two matrices supplied by you. Usually,

Q

and

Z

are chosen as the orthogonal matrices returned by f08xec. Equivalently,

Q

and

Z

are the left and right Schur vectors of the matrix pair supplied to f08xec. In that case,

Q Y

and

Z X

are the left and right generalized eigenvectors, respectively, of the matrix pair supplied to f08xec.

A

must be block upper triangular; with

1 \times 1

and

2 \times 2

diagonal blocks. Corresponding to each

2 \times 2

diagonal block is a complex conjugate pair of eigenvalues and eigenvectors; only one eigenvector of the pair is computed, namely the one corresponding to the eigenvalue with positive imaginary part. Each

1 \times 1

block gives a real generalized eigenvalue and a corresponding eigenvector.

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

Golub G H and Van Loan C F (1996) Matrix Computations (3rd Edition) Johns Hopkins University Press, Baltimore

Moler C B and Stewart G W (1973) An algorithm for generalized matrix eigenproblems SIAM J. Numer. Anal. 10 241–256

Stewart G W and Sun J-G (1990) Matrix Perturbation Theory Academic Press, London

5 Arguments

1: $order$ – Nag_OrderType Input

On entry: the order argument specifies the two-dimensional storage scheme being used, i.e., row-major ordering or column-major ordering. C language defined storage is specified by

order = Nag_RowMajor

. See Section 3.1.3 in the Introduction to the NAG Library CL Interface for a more detailed explanation of the use of this argument.

Constraint:

order = Nag_RowMajor

Nag_ColMajor

2: $side$ – Nag_SideType Input

On entry: specifies the required sets of generalized eigenvectors.

$side = Nag_RightSide$: Only right eigenvectors are computed.
$side = Nag_LeftSide$: Only left eigenvectors are computed.
$side = Nag_BothSides$: Both left and right eigenvectors are computed.

Constraint:

side = Nag_BothSides

Nag_LeftSide

Nag_RightSide

3: $how_many$ – Nag_HowManyType Input

On entry: specifies further details of the required generalized eigenvectors.

$how_many = Nag_ComputeAll$: All right and/or left eigenvectors are computed.
$how_many = Nag_BackTransform$: All right and/or left eigenvectors are computed; they are backtransformed using the input matrices supplied in arrays vr and/or vl.
$how_many = Nag_ComputeSelected$: Selected right and/or left eigenvectors, defined by the array select, are computed.

Constraint:

how_many = Nag_ComputeAll

Nag_BackTransform

Nag_ComputeSelected

4: $select [\dim]$ – const Nag_Boolean Input

Note: the dimension, dim, of the array select must be at least

$n$ when $how_many = Nag_ComputeSelected$ ;
otherwise select may be NULL.

On entry: specifies the eigenvectors to be computed if

how_many = Nag_ComputeSelected

. To select the generalized eigenvector corresponding to the

j

th generalized eigenvalue, the

j

th element of select should be set to Nag_TRUE; if the eigenvalue corresponds to a complex conjugate pair, then real and imaginary parts of eigenvectors corresponding to the complex conjugate eigenvalue pair will be computed.

how_many = Nag_ComputeAll

Nag_BackTransform

, select is not referenced and may be NULL.

Constraint: if

how_many = Nag_ComputeSelected

select [j] = Nag_TRUE

Nag_FALSE

, for

j = 0, 1, \dots, n - 1

5: $n$ – Integer Input

On entry:

n

, the order of the matrices

A

and

B

Constraint:

n \geq 0

6: $a [\dim]$ – const double Input

Note: the dimension, dim, of the array a must be at least

pda \times n

The

(i, j)

th element of the matrix

A

is stored in

$a [(j - 1) \times pda + i - 1]$ when $order = Nag_ColMajor$ ;
$a [(i - 1) \times pda + j - 1]$ when $order = Nag_RowMajor$ .

On entry: the matrix pair

(A, B)

must be in the generalized Schur form. Usually, this is the matrix

A

returned by f08xec.

7: $pda$ – Integer Input

On entry: the stride separating row or column elements (depending on the value of order) in the array a.

Constraint:

pda \geq \max (1, n)

8: $b [\dim]$ – const double Input

Note: the dimension, dim, of the array b must be at least

pdb \times n

The

(i, j)

th element of the matrix

B

is stored in

$b [(j - 1) \times pdb + i - 1]$ when $order = Nag_ColMajor$ ;
$b [(i - 1) \times pdb + j - 1]$ when $order = Nag_RowMajor$ .

On entry: the matrix pair

(A, B)

must be in the generalized Schur form. If

A

has a

2 \times 2

diagonal block then the corresponding

2 \times 2

block of

B

must be diagonal with positive elements. Usually, this is the matrix

B

returned by f08xec.

9: $pdb$ – Integer Input

On entry: the stride separating row or column elements (depending on the value of order) in the array b.

Constraint:

pdb \geq \max (1, n)

10: $vl [\dim]$ – double Input/Output

Note: the dimension, dim, of the array vl must be at least

$pdvl \times mm$ when $side = Nag_LeftSide$ or $Nag_BothSides$ and $order = Nag_ColMajor$ ;
$n \times pdvl$ when $side = Nag_LeftSide$ or $Nag_BothSides$ and $order = Nag_RowMajor$ ;
otherwise vl may be NULL.

The

(i, j)

th element of the matrix is stored in

$vl [(j - 1) \times pdvl + i - 1]$ when $order = Nag_ColMajor$ ;
$vl [(i - 1) \times pdvl + j - 1]$ when $order = Nag_RowMajor$ .

On entry: if

how_many = Nag_BackTransform

and

side = Nag_LeftSide

Nag_BothSides

, vl must be initialized to an

n \times n

matrix

Q

. Usually, this is the orthogonal matrix

Q

of left Schur vectors returned by f08xec.

On exit: if

side = Nag_LeftSide

Nag_BothSides

, vl contains:

if $how_many = Nag_ComputeAll$ , the matrix $Y$ of left eigenvectors of $(A, B)$ ;
if $how_many = Nag_BackTransform$ , the matrix $Q Y$ ;
if $how_many = Nag_ComputeSelected$ , the left eigenvectors of $(A, B)$ specified by select, stored consecutively in the rows or columns (depending on the value of order) of the array vl, in the same order as their corresponding eigenvalues.

A complex eigenvector corresponding to a complex eigenvalue is stored in two consecutive rows or columns, the first holding the real part, and the second the imaginary part.

side = Nag_RightSide

, vl is not referenced and may be NULL.

11: $pdvl$ – Integer Input

On entry: the stride separating row or column elements (depending on the value of order) in the array vl.

Constraints:

if $order = Nag_ColMajor$ ,
- if $side = Nag_LeftSide$ or $Nag_BothSides$ , $pdvl \geq n$ ;
- if $side = Nag_RightSide$ , vl may be NULL;
if $order = Nag_RowMajor$ ,
- if $side = Nag_LeftSide$ or $Nag_BothSides$ , $pdvl \geq mm$ ;
- if $side = Nag_RightSide$ , vl may be NULL.

12: $vr [\dim]$ – double Input/Output

Note: the dimension, dim, of the array vr must be at least

$pdvr \times mm$ when $side = Nag_RightSide$ or $Nag_BothSides$ and $order = Nag_ColMajor$ ;
$n \times pdvr$ when $side = Nag_RightSide$ or $Nag_BothSides$ and $order = Nag_RowMajor$ ;
otherwise vr may be NULL.

The

(i, j)

th element of the matrix is stored in

$vr [(j - 1) \times pdvr + i - 1]$ when $order = Nag_ColMajor$ ;
$vr [(i - 1) \times pdvr + j - 1]$ when $order = Nag_RowMajor$ .

On entry: if

how_many = Nag_BackTransform

and

side = Nag_RightSide

Nag_BothSides

, vr must be initialized to an

n \times n

matrix

Z

. Usually, this is the orthogonal matrix

Z

of right Schur vectors returned by f08xec.

On exit: if

side = Nag_RightSide

Nag_BothSides

, vr contains:

if $how_many = Nag_ComputeAll$ , the matrix $X$ of right eigenvectors of $(A, B)$ ;
if $how_many = Nag_BackTransform$ , the matrix $Z X$ ;
if $how_many = Nag_ComputeSelected$ , the right eigenvectors of $(A, B)$ specified by select, stored consecutively in the rows or columns (depending on the value of order) of the array vr, in the same order as their corresponding eigenvalues.

A complex eigenvector corresponding to a complex eigenvalue is stored in two consecutive rows or columns, the first holding the real part, and the second the imaginary part.

side = Nag_LeftSide

, vr is not referenced and may be NULL.

13: $pdvr$ – Integer Input

On entry: the stride separating row or column elements (depending on the value of order) in the array vr.

Constraints:

if $order = Nag_ColMajor$ ,
- if $side = Nag_RightSide$ or $Nag_BothSides$ , $pdvr \geq n$ ;
- if $side = Nag_LeftSide$ , vr may be NULL;
if $order = Nag_RowMajor$ ,
- if $side = Nag_RightSide$ or $Nag_BothSides$ , $pdvr \geq mm$ ;
- if $side = Nag_LeftSide$ , vr may be NULL.

14: $mm$ – Integer Input

On entry: the number of columns in the arrays vl and/or vr.

Constraints:

if $how_many = Nag_ComputeAll$ or $Nag_BackTransform$ , $mm \geq n$ ;
if $how_many = Nag_ComputeSelected$ , mm must not be less than the number of requested eigenvectors.

15: $m$ – Integer * Output

On exit: the number of columns in the arrays vl and/or vr actually used to store the eigenvectors. If

how_many = Nag_ComputeAll

Nag_BackTransform

, m is set to n. Each selected real eigenvector occupies one row or column and each selected complex eigenvector occupies two rows or columns.

16: $fail$ – NagError * Input/Output

The NAG error argument (see Section 7 in the Introduction to the NAG Library CL Interface).

6 Error Indicators and Warnings

NE_ALLOC_FAIL: Dynamic memory allocation failed.
See Section 3.1.2 in the Introduction to the NAG Library CL Interface for further information.
NE_BAD_PARAM: On entry, argument $⟨ value ⟩$ had an illegal value.
NE_CONSTRAINT: Constraint: if $how_many = Nag_ComputeSelected$ , $select [j] = Nag_TRUE$ or $Nag_FALSE$ , for $j = 0, 1, \dots, n - 1$ .
NE_ENUM_INT_2: On entry, $how_many = ⟨ value ⟩$ , $n = ⟨ value ⟩$ and $mm = ⟨ value ⟩$ .
Constraint: if $how_many = Nag_ComputeAll$ or $Nag_BackTransform$ , $mm \geq n$ ;
if $how_many = Nag_ComputeSelected$ , mm must not be less than the number of requested eigenvectors.

On entry, $side = ⟨ value ⟩$ , $pdvl = ⟨ value ⟩$ and $mm = ⟨ value ⟩$ .
Constraint: if $side = Nag_LeftSide$ or $Nag_BothSides$ , $pdvl \geq mm$ .

On entry, $side = ⟨ value ⟩$ , $pdvl = ⟨ value ⟩$ and $n = ⟨ value ⟩$ .
Constraint: if $side = Nag_LeftSide$ or $Nag_BothSides$ , $pdvl \geq n$ .

On entry, $side = ⟨ value ⟩$ , $pdvr = ⟨ value ⟩$ and $mm = ⟨ value ⟩$ .
Constraint: if $side = Nag_RightSide$ or $Nag_BothSides$ , $pdvr \geq mm$ .

On entry, $side = ⟨ value ⟩$ , $pdvr = ⟨ value ⟩$ and $n = ⟨ value ⟩$ .
Constraint: if $side = Nag_RightSide$ or $Nag_BothSides$ , $pdvr \geq n$ .
NE_INT: On entry, $n = ⟨ value ⟩$ .
Constraint: $n \geq 0$ .

On entry, $pda = ⟨ value ⟩$ .
Constraint: $pda > 0$ .

On entry, $pdb = ⟨ value ⟩$ .
Constraint: $pdb > 0$ .

On entry, $pdvl = ⟨ value ⟩$ .
Constraint: $pdvl > 0$ .

On entry, $pdvr = ⟨ value ⟩$ .
Constraint: $pdvr > 0$ .
NE_INT_2: On entry, $pda = ⟨ value ⟩$ and $n = ⟨ value ⟩$ .
Constraint: $pda \geq \max (1, n)$ .

On entry, $pdb = ⟨ value ⟩$ and $n = ⟨ value ⟩$ .
Constraint: $pdb \geq \max (1, n)$ .
NE_INTERNAL_ERROR: An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
See Section 7.5 in the Introduction to the NAG Library CL Interface for further information.
NE_NO_LICENCE: Your licence key may have expired or may not have been installed correctly.
See Section 8 in the Introduction to the NAG Library CL Interface for further information.
NE_NOT_COMPLEX: The $2 \times 2$ block $(⟨ value ⟩ : ⟨ value ⟩ + 1)$ does not have complex eigenvalues.

7 Accuracy

It is beyond the scope of this manual to summarise the accuracy of the solution of the generalized eigenvalue problem. Interested readers should consult Section 4.11 of the LAPACK Users' Guide (see Anderson et al. (1999)) and Chapter 6 of Stewart and Sun (1990).

8 Parallelism and Performance

Background information to multithreading can be found in the Multithreading documentation.

f08ykc 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 function. Please also consult the Users' Note for your implementation for any additional implementation-specific information.

9 Further Comments

f08ykc is the sixth step in the solution of the real generalized eigenvalue problem and is called after f08xec.

The complex analogue of this function is f08yxc.

10 Example

This example computes the

α

and

β

arguments, which defines the generalized eigenvalues and the corresponding left and right eigenvectors, of the matrix pair

(A, B)

given by

A = (\begin{array}{r} 1.0 & 1.0 & 1.0 & 1.0 & 1.0 \\ 2.0 & 4.0 & 8.0 & 16.0 & 32.0 \\ 3.0 & 9.0 & 27.0 & 81.0 & 243.0 \\ 4.0 & 16.0 & 64.0 & 256.0 & 1024.0 \\ 5.0 & 25.0 & 125.0 & 625.0 & 3125.0 \end{array}) and B = (\begin{array}{r} 1.0 & 2.0 & 3.0 & 4.0 & 5.0 \\ 1.0 & 4.0 & 9.0 & 16.0 & 25.0 \\ 1.0 & 8.0 & 27.0 & 64.0 & 125.0 \\ 1.0 & 16.0 & 81.0 & 256.0 & 625.0 \\ 1.0 & 32.0 & 243.0 & 1024.0 & 3125.0 \end{array}) .

To compute generalized eigenvalues, it is required to call five functions: f08whc to balance the matrix, f08aec to perform the

Q R

factorization of

B

, f08agc to apply

Q

A

, f08wfc to reduce the matrix pair to the generalized Hessenberg form and f08xec to compute the eigenvalues via the

Q Z

algorithm.

The computation of generalized eigenvectors is done by calling f08ykc to compute the eigenvectors of the balanced matrix pair. The function f08wjc is called to backward transform the eigenvectors to the user-supplied matrix pair. If both left and right eigenvectors are required then f08wjc must be called twice.

f08yk: FL CL CPP AD PY MB

NAG CL Interfacef08ykc (dtgevc)

▸▿ Contents

1 Purpose

2 Specification

3 Description

4 References

5 Arguments

6 Error Indicators and Warnings

7 Accuracy

8 Parallelism and Performance

9 Further Comments

10 Example

10.1 Program Text

10.2 Program Data

10.3 Program Results

NAG CL Interface
f08ykc (dtgevc)