next up previous contents index
Next: Driver Routines for Singular Up: Generalized Nonsymmetric Eigenvalue Problems Previous: Arguments   Contents   Index

Example (from Program LA_GGEVX_EXAMPLE)

The results below are computed with $\epsilon = 1.19209 \times 10^{-7}$.

\begin{displaymath}
A = \left( \begin{array}{rrrrr}
-2 - 9i &
1 + 2i &
-6 +2i &
...
... - 8i &
5 + 10i &
-4 + 2i &
1 -i &
-4 + 0i
\end{array} \right)
\end{displaymath}


\begin{displaymath}
B = \left( \begin{array}{rrrrr}
-7 + 5i &
4 + 12i &
5 + ...
...
-2 - 2i &
-10 - 3i &
4 + 3i &
-3 -i
\end{array} \right)
\end{displaymath}

Arrays A and B on entry:

\begin{displaymath}
\begin{array}{c} {\bf A} \\
\begin{array}{\vert rrrrr\vert}...
...&
(-4, 2) &
(1, -1)&
(-4, 0) \\
\hline \end{array} \end{array}\end{displaymath}


\begin{displaymath}
\begin{array}{c} {\bf B} \\
\begin{array}{\vert rrrrr\vert}...
...(-10, -3) &
(4, 3)&
(-3, -1)
\\
\hline \end{array} \end{array}\end{displaymath}

The call:


   CALL LA_GGEVX( A, B, ALPHA, BETA, BALANC='B', LSCALE=LSCALE, & 

RSCALE=RSCALE, ABNRM=ABNRM, &
BBNRM=BBNRM, RCONDE=RCONDE, &
RCONDV=RCONDV )
${\bf LSCALE, RSCALE, ABNRM, BBNRM, RCONDE, RCONDV, ALPHA, BETA}$ and LAMBDA on exit:

\begin{displaymath}\hspace{-0.50 cm} \begin{array}{cc} {\bf LSCALE} \\
\begin{...
...1.00000 & 1.00000 & 1.00000 \\
\hline \end{array} \end{array} \end{displaymath}


\begin{displaymath}\begin{array}{cc}
{\bf ABNRM} = 2.76734 \times 10^{1} \hspace{1.00 cm} {\bf BBNRM} = 3.61864 \times 10^{1}
\end{array} \end{displaymath}


\begin{displaymath}\hspace{-0.50 cm} \begin{array}{cc} {\bf RCONDE} \\
\begin{...
... 2.67197 & 1.45866 & 1.61397\\
\hline \end{array} \end{array} \end{displaymath}


\begin{displaymath}\hspace{-0.50 cm} \begin{array}{cc} {\bf ALPHA} \\
\begin{a...
....000, 0.000)& (-1.509,2.026) \\
\hline \end{array} \end{array}\end{displaymath}


\begin{displaymath}\hspace{-0.50 cm} \begin{array}{cc} {\bf BETA} \\
\begin{ar...
...3.794,0.000)& (17.099,0.000) \\
\hline \end{array} \end{array}\end{displaymath}


\begin{displaymath}\hspace{-0.50 cm} \begin{array}{cc} {\bf LAMBDA} \\
\begin{...
...0.000,0.000)& (-0.088,0.119) \\
\hline \end{array} \end{array}\end{displaymath}

Balancing was not needed.

The $l_{1}$ norm of $A$ is $2.76734 \times 10^{1}$ and the $l_{1}$ norm of $B$ is $3.61864 \times 10^{1}$.

The reciprocal condition numbers of the eigenvalues are:

\begin{displaymath}\left( \begin{array}{lllll}
2.12830 & 1.84798 & 8.09452 & 5.56294 & 4.77657
\end{array} \right) \end{displaymath}

The reciprocal condition numbers of the eigenvectors are:

\begin{displaymath}\left( \begin{array}{lllll}
2.29658 & 1.83061 & 2.67197 & 1.45866 & 1.61397
\end{array} \right) \end{displaymath}


next up previous contents index
Next: Driver Routines for Singular Up: Generalized Nonsymmetric Eigenvalue Problems Previous: Arguments   Contents   Index
Susan Blackford 2001-08-19