F01 Chapter Contents
F01 Chapter Introduction
NAG Library Manual

# NAG Library Routine DocumentF01KCF

Note:  before using this routine, please read the Users' Note for your implementation to check the interpretation of bold italicised terms and other implementation-dependent details.

## 1  Purpose

F01KCF computes an estimate of the absolute condition number of a matrix function $f$ of a complex $n$ by $n$ matrix $A$ in the $1$-norm, using analytical derivatives of $f$ you have supplied.

## 2  Specification

 SUBROUTINE F01KCF ( N, A, LDA, F, IUSER, RUSER, IFLAG, CONDA, NORMA, NORMFA, IFAIL)
 INTEGER N, LDA, IUSER(*), IFLAG, IFAIL REAL (KIND=nag_wp) RUSER(*), CONDA, NORMA, NORMFA COMPLEX (KIND=nag_wp) A(LDA,*) EXTERNAL F

## 3  Description

The absolute condition number of $f$ at $A$, ${\mathrm{cond}}_{\mathrm{abs}}\left(f,A\right)$ is given by the norm of the Fréchet derivative of $f$, $L\left(A\right)$, which is defined by
 $LX := maxE≠0 LX,E E ,$
where $L\left(X,E\right)$ is the Fréchet derivative in the direction $E$. $L\left(X,E\right)$ is linear in $E$ and can therefore be written as
 $vec LX,E = KX vecE ,$
where the $\mathrm{vec}$ operator stacks the columns of a matrix into one vector, so that $K\left(X\right)$ is ${n}^{2}×{n}^{2}$. F01KCF computes an estimate $\gamma$ such that $\gamma \le {‖K\left(X\right)‖}_{1}$, where ${‖K\left(X\right)‖}_{1}\in \left[{n}^{-1}{‖L\left(X\right)‖}_{1},n{‖L\left(X\right)‖}_{1}\right]$. The relative condition number can then be computed via
 $cond rel f,A = cond abs f,A A1 fA 1 .$
The algorithm used to find $\gamma$ is detailed in Section 3.4 of Higham (2008).

## 4  References

Higham N J (2008) Functions of Matrices: Theory and Computation SIAM, Philadelphia, PA, USA

## 5  Parameters

1:     $\mathrm{N}$ – INTEGERInput
On entry: $n$, the order of the matrix $A$.
Constraint: ${\mathbf{N}}\ge 0$.
2:     $\mathrm{A}\left({\mathbf{LDA}},*\right)$ – COMPLEX (KIND=nag_wp) arrayInput/Output
Note: the second dimension of the array A must be at least ${\mathbf{N}}$.
On entry: the $n$ by $n$ matrix $A$.
On exit: the $n$ by $n$ matrix, $f\left(A\right)$.
3:     $\mathrm{LDA}$ – INTEGERInput
On entry: the first dimension of the array A as declared in the (sub)program from which F01KCF is called.
Constraint: ${\mathbf{LDA}}\ge {\mathbf{N}}$.
4:     $\mathrm{F}$ – SUBROUTINE, supplied by the user.External Procedure
Given an integer $m$, the subroutine F evaluates ${f}^{\left(m\right)}\left({z}_{i}\right)$ at a number of points ${z}_{i}$.
The specification of F is:
 SUBROUTINE F ( M, IFLAG, NZ, Z, FZ, IUSER, RUSER)
 INTEGER M, IFLAG, NZ, IUSER(*) REAL (KIND=nag_wp) RUSER(*) COMPLEX (KIND=nag_wp) Z(NZ), FZ(NZ)
1:     $\mathrm{M}$ – INTEGERInput
On entry: the order, $m$, of the derivative required.
If ${\mathbf{M}}=0$, $f\left({z}_{i}\right)$ should be returned. For ${\mathbf{M}}>0$, ${f}^{\left(m\right)}\left({z}_{i}\right)$ should be returned.
2:     $\mathrm{IFLAG}$ – INTEGERInput/Output
On entry: IFLAG will be zero.
On exit: IFLAG should either be unchanged from its entry value of zero, or may be set nonzero to indicate that there is a problem in evaluating the function $f\left(z\right)$; for instance $f\left(z\right)$ may not be defined. If IFLAG is returned as nonzero then F01KCF will terminate the computation, with ${\mathbf{IFAIL}}={\mathbf{3}}$.
3:     $\mathrm{NZ}$ – INTEGERInput
On entry: ${n}_{z}$, the number of function or derivative values required.
4:     $\mathrm{Z}\left({\mathbf{NZ}}\right)$ – COMPLEX (KIND=nag_wp) arrayInput
On entry: the ${n}_{z}$ points ${z}_{1},{z}_{2},\dots ,{z}_{{n}_{z}}$ at which the function $f$ is to be evaluated.
5:     $\mathrm{FZ}\left({\mathbf{NZ}}\right)$ – COMPLEX (KIND=nag_wp) arrayOutput
On exit: the ${n}_{z}$ function or derivative values. ${\mathbf{FZ}}\left(\mathit{i}\right)$ should return the value ${f}^{\left(m\right)}\left({z}_{\mathit{i}}\right)$, for $\mathit{i}=1,2,\dots ,{n}_{z}$.
6:     $\mathrm{IUSER}\left(*\right)$ – INTEGER arrayUser Workspace
7:     $\mathrm{RUSER}\left(*\right)$ – REAL (KIND=nag_wp) arrayUser Workspace
F is called with the parameters IUSER and RUSER as supplied to F01KCF. You are free to use the arrays IUSER and RUSER to supply information to F as an alternative to using COMMON global variables.
F must either be a module subprogram USEd by, or declared as EXTERNAL in, the (sub)program from which F01KCF is called. Parameters denoted as Input must not be changed by this procedure.
5:     $\mathrm{IUSER}\left(*\right)$ – INTEGER arrayUser Workspace
6:     $\mathrm{RUSER}\left(*\right)$ – REAL (KIND=nag_wp) arrayUser Workspace
IUSER and RUSER are not used by F01KCF, but are passed directly to F and may be used to pass information to this routine as an alternative to using COMMON global variables.
7:     $\mathrm{IFLAG}$ – INTEGEROutput
On exit: ${\mathbf{IFLAG}}=0$, unless IFLAG has been set nonzero inside F, in which case IFLAG will be the value set and IFAIL will be set to ${\mathbf{IFAIL}}={\mathbf{3}}$.
8:     $\mathrm{CONDA}$ – REAL (KIND=nag_wp)Output
On exit: an estimate of the absolute condition number of $f$ at $A$.
9:     $\mathrm{NORMA}$ – REAL (KIND=nag_wp)Output
On exit: the $1$-norm of $A$.
10:   $\mathrm{NORMFA}$ – REAL (KIND=nag_wp)Output
On exit: the $1$-norm of $f\left(A\right)$.
11:   $\mathrm{IFAIL}$ – INTEGERInput/Output
On entry: IFAIL must be set to $0$, $-1\text{​ or ​}1$. If you are unfamiliar with this parameter you should refer to Section 3.3 in the Essential Introduction for details.
For environments where it might be inappropriate to halt program execution when an error is detected, the value $-1\text{​ or ​}1$ is recommended. If the output of error messages is undesirable, then the value $1$ is recommended. Otherwise, if you are not familiar with this parameter, the recommended value is $0$. When the value $-\mathbf{1}\text{​ or ​}\mathbf{1}$ is used it is essential to test the value of IFAIL on exit.
On exit: ${\mathbf{IFAIL}}={\mathbf{0}}$ unless the routine detects an error or a warning has been flagged (see Section 6).

## 6  Error Indicators and Warnings

If on entry ${\mathbf{IFAIL}}={\mathbf{0}}$ or $-{\mathbf{1}}$, explanatory error messages are output on the current error message unit (as defined by X04AAF).
Errors or warnings detected by the routine:
${\mathbf{IFAIL}}=1$
An internal error occurred when estimating the norm of the Fréchet derivative of $f$ at $A$. Please contact NAG.
${\mathbf{IFAIL}}=2$
An internal error occurred when evaluating the matrix function $f\left(A\right)$. You can investigate further by calling F01FMF with the matrix $A$ and the function $f$.
${\mathbf{IFAIL}}=3$
IFLAG has been set nonzero by the user-supplied subroutine.
${\mathbf{IFAIL}}=-1$
On entry, ${\mathbf{N}}<0$.
${\mathbf{IFAIL}}=-3$
On entry, parameter LDA is invalid.
Constraint: ${\mathbf{LDA}}\ge {\mathbf{N}}$.
${\mathbf{IFAIL}}=-99$
See Section 3.8 in the Essential Introduction for further information.
${\mathbf{IFAIL}}=-399$
Your licence key may have expired or may not have been installed correctly.
See Section 3.7 in the Essential Introduction for further information.
${\mathbf{IFAIL}}=-999$
Dynamic memory allocation failed.
See Section 3.6 in the Essential Introduction for further information.

## 7  Accuracy

F01KCF uses the norm estimation routine F04ZDF to estimate a quantity $\gamma$, where $\gamma \le {‖K\left(X\right)‖}_{1}$ and ${‖K\left(X\right)‖}_{1}\in \left[{n}^{-1}{‖L\left(X\right)‖}_{1},n{‖L\left(X\right)‖}_{1}\right]$. For further details on the accuracy of norm estimation, see the documentation for F04ZDF.

## 8  Parallelism and Performance

F01KCF is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library. In these implementations, this routine may make calls to the user-supplied functions from within an OpenMP parallel region. Thus OpenMP directives within the user functions can only be used if you are compiling the user-supplied function and linking the executable in accordance with the instructions in the Users' Note for your implementation. The user workspace arrays IUSER and RUSER are classified as OpenMP shared memory and use of IUSER and RUSER has to take account of this in order to preserve thread safety whenever information is written back to either of these arrays. If at all possible, it is recommended that these arrays are only used to supply read-only data to the user functions when a multithreaded implementation is being used.
F01KCF 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.

Approximately $6{n}^{2}$ of complex allocatable memory is required by the routine, in addition to the memory used by the underlying matrix function routine F01FMF.
F01KCF returns the matrix function $f\left(A\right)$. This is computed using F01FMF. If only $f\left(A\right)$ is required, without an estimate of the condition number, then it is far more efficient to use F01FMF directly.
The real analogue of this routine is F01JCF.

## 10  Example

This example estimates the absolute and relative condition numbers of the matrix function ${e}^{3A}$ where
 $A= 1.0+1.0i 0.0+1.0i 1.0+0.0i 2.0+0.0i 0.0+0.0i 2.0+0.0i 0.0+2.0i 1.0+0.0i 0.0+1.0i 0.0+1.0i 0.0+0.0i 2.0+0.0i 1.0+0.0i 0.0+1.0i 1.0+0.0i 0.0+1.0i .$

### 10.1  Program Text

Program Text (f01kcfe.f90)

### 10.2  Program Data

Program Data (f01kcfe.d)

### 10.3  Program Results

Program Results (f01kcfe.r)