Hi!
Define A = $$\begin{pmatrix}
3 & 3\\
-3 & -3\\
\end{pmatrix}$$
Find a matrix such that
$$P^{-1}AP = \begin{pmatrix}
\lambda & 1\\
0 & \lambda\\
\end{pmatrix}$$
Where lambda is the unique eigenvalue of A.
What I have is;
the characteristic equation is given by
$$(3-\lambda)(-3-\lambda)+9 = 0\\
\lambda^2+3\lambda-3\lambda-9+9\\
\lambda^2 = 0\\
\lambda = 0\\$$
So we have $$\lambda = 0$$ with $$am(\lambda) = 2$$
To find an eigenvector we have
$$\begin{pmatrix}
3 & 3\\
-3 & -3\\
\end{pmatrix} \begin{pmatrix}
\mu_{11} \\
\mu_{12} \\
\end{pmatrix}
= 0
\begin{pmatrix}
\mu_{11} \\
\mu_{12} \\
\end{pmatrix} \\
3\mu_{11}+3\mu_{12} = 0\\
\mbox{and} \\
-3\mu_{11}-3\mu_{12} = 0\\$$
Pick $$\mu_{11} = 1\\
\mbox{then}\\
\mu_{12} = -1\\
\mbox{and}\\
\mu_1 = \begin{pmatrix}
1 \\
-1
\end{pmatrix}$$
Now normally I'd find a generalized eigenvector using
$$\begin{pmatrix}
3-\lambda & 3\\
-3 & -3-\lambda\\
\end{pmatrix}$$
But in this case this gives
$$\begin{pmatrix}
3-0 & 3\\
-3 & -3-0\\
\end{pmatrix}$$
Which doesn't really lead me anywhere.
How do I find another eigenvector such that I can use
$$P = \begin{pmatrix}
\mu_1 & \mu_2\\
\end{pmatrix}$$
Reinout
Yeah that does help!
How foolish of me not to think of it in that way.
However I do find it suprising to see the $$\begin{pmatrix}
0\\
0\\
\end{pmatrix}$$ vector as an eigenvector since that eigenvector is basically applicable to any matrix. Nevertheless I wouldn't know of any other way to solve this.
Thanks Alan.