2
$\begingroup$

Find the Generalized Eigenvectors of
$$ \begin{bmatrix} 1 & 0 & 0 & 0 & 0\\ 1 & -1 & 0 & 0 & -1\\ 1 & -1 & 0 & 0 & -1\\ 0 & 0 & 0 & 0 & -1\\ -1 & 1 & 0 & 0 & 1 \end{bmatrix} $$

Since for $\lambda = 0, null(A-\lambda I) = 2.$ I have 2 eigenvectors = $$ \begin{bmatrix} 0\\0\\0\\1\\0 \end{bmatrix} and \begin{bmatrix} 0\\0\\1\\0\\0 \end{bmatrix} $$ I know to use $(A - \lambda I)v_2 = v_1$ to get all eigenvectors corresponding to a $\lambda$. But this one has 2 blocks corresponding to $\lambda = 0$ in the Jordan Cannonical form. $$ J = \begin{bmatrix} 1 & 0 & 0 & 0 & 0\\ 0 & 0 & 0 & 0 & 0\\ 0 & 0 & 0 & 1 & 0\\ 0 & 0 & 0 & 0 & 1\\ 0 & 0 & 0 & 0 & 0 \end{bmatrix} $$ So, how do I associate generalised eigenvectors to each Jordan block?

$\endgroup$
  • $\begingroup$ What is generalised eigenvector? $\endgroup$ – Mostafa Ayaz Jan 29 '18 at 18:44
  • $\begingroup$ Assuming that you know how to calculate general eigenvectors, how long is the Jordan chain for each of the eigenvectors of zero that you’ve got? $\endgroup$ – amd Jan 30 '18 at 1:06
  • $\begingroup$ For eigenvalue = 0 it's 1 and 3, For eigenvalue = 1 it's 1 $\endgroup$ – Renganathan Subramanian Jan 30 '18 at 9:52
1
$\begingroup$

The first Jordan block is the one-by-one matrix [1]. An associated eigenvector is x = [–1, –1, –1, –1, 1]$^{\mathrm T}.$ The multiplicity of that eigenvalue is one, so there are no complications here.

Dealing with eigenvalue 0 is not as easy. The remainder of my answer is based on Wikipedia subtopic "Computation of generalized eigenvectors."

Because rank(A – 0 I) = rank A = 3, rank A$^2$ = 2, and rank A$^3$ = 1, there will be three linearly-independent generalized eigenvectors—z$_3$ of order 3, z$_2$ of order 2, and z$_1$ of order 1—that form a chain. To find z$_3,$ we solve A$^3$z$_3$ = 0 such that A$^2$z$_3$ ≠ 0. The simplest solution is z$_3$ = [0, 0, 0, 0, 1]$^{\mathrm T}.$ Then z$_2$ = Az$_3$ = [0, –1, –1, –1, 1]$^{\mathrm T},$ and z$_1$ = Az$_2$ = [0, 0, 0, –1, 0]$^{\mathrm T}.$ Those three generalized eigenvectors are associated with the third, three-by-three Jordan block.

Finally, note that z$_1$ is a multiple of the first eigenvector that you give. Hence, you can associate the other eigenvector y = [0, 0, 1, 0, 0]$^{\mathrm T}$ with the second, one-by-one Jordan block [0].

FYI, the Jordan decomposition A = S J S$^{-1}$ has S = [x, y, z$_1,$ z$_2,$ z$_3]$ where the generalized eigenvectors are the columns of S.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.