A = $\begin{bmatrix} 1 & -1 & 0 & -1\\ 0 & 2 & 0 & 1\\ -2 & 1 & -1 & 1\\ 2 & -1 & 2 & 0\\ \end{bmatrix}$
I've been given this matrix - I'm supposed to find its Jordan form as well as matrix M such that: $ A = MJM^{-1}$.
So far, what I have is the eigenvalues: 1, 1, 1, -1. I have three eigenvectors (which I realize is not enough):
A-1I ~ Gauss Jordan ~ $\begin{bmatrix} 1 & 0 & 1 & 0\\ 0 & 1 & 0 & 1\\ 0 & 0 & 0 & 0\\ 0 & 0 & 0 & 0\\ \end{bmatrix}$ $ \qquad \Rightarrow v_1 = \begin{pmatrix} -1 \\ 0 \\ 1\\ 0\\ \end{pmatrix}$ $ \quad v_2 = \begin{pmatrix} 0 \\ -1 \\ 0\\ 1\\ \end{pmatrix}$
A+1I ~~ $\begin{bmatrix} 1 & 1 & 0 & 0\\ 0 & 3 & 0 & 1\\ 0 & 0 & 1 & 1\\ 0 & 0 & 0 & 0\\ \end{bmatrix}$ $ \qquad \Rightarrow v_3 = \begin{pmatrix} 1 \\ -1 \\ -3\\ 3\\ \end{pmatrix}$
This is all I've been able to find. I need to either find a fourth vector in order to find matrix M, or a way to find matrix J and then compute M.
If you had four eigenvectors, you'd be able to diagonalize the matrix. Evidently this matrix is not diagonalizable, but of course it still has a Jordan form. You need to find the "generalized eigenspaces", i.e. the kernels of not just the $(A-\lambda I)$, but also $(A-\lambda I)^n$ for all $n \geq 1$. Since you're only missing one generalized eigenvector here, you'll only need to look at $n=2$ to find your $v_4$.