2
$\begingroup$

Let $A$ be an $n\times n$ integral matrix and $\xi$ an integral vector of dimension $n$. Let $$W_i=[\xi,A\xi,\ldots,A^{i-1}\xi]$$ for each positive $i$. Let $d_{i,j}$ (where $1\le j\le i$) denote the $j$-th invariant factor of $W_i$. It seems that $d_{i,j}$ only depends on $j$, that is $d_{j,j}=d_{j+1,j}=\cdots=d_{n,j}=d_{n+1,j}=\cdots$. By Cayley-Hamilton Theorem, we know that $W_i$ is equivalent (in $\mathbb{Z}$) to $[W_n,0_{n\times (i-n)}]$ for $i>n$. Thus we only need to consider the case that $i\le n$.

For example, let \begin{equation*} A=\left(\begin{matrix} 2&2&3&3\\ 1&3&0&0\\ 3&3&0&3\\ 0&0&3&2\\ \end{matrix} \right),\quad \xi=\left(\begin{matrix} 1\\ 2\\ 3\\ 0\\ \end{matrix} \right) \end{equation*} Then \begin{equation*} W_1=\left(\begin{matrix} 1\\ 2\\ 3\\ 0\\ \end{matrix} \right), W_2=\left(\begin{matrix} 1&15\\ 2&7\\ 3&9\\ 0&9\\ \end{matrix} \right),W_3=\left(\begin{matrix} 1&15&98\\ 2&7&36\\ 3&9&93\\ 0&9&45\\ \end{matrix} \right), W_4=\left(\begin{matrix} 1&15&98&682\\ 2&7&36&206\\ 3&9&93&537\\ 0&9&45&369\\ \end{matrix} \right) \end{equation*} The Smith normal forms of these matrices are \begin{equation*} \left(\begin{matrix} 1\\ 0\\ 0\\ 0\\ \end{matrix} \right), \left(\begin{matrix} 1&0\\ 0&1\\ 0&0\\ 0&0\\ \end{matrix} \right),\left(\begin{matrix} 1&0&0\\ 0&1&0\\ 0&0&3\\ 0&0&0\\ \end{matrix} \right), \left(\begin{matrix} 1&0&0&0\\ 0&1&0&0\\ 0&0&3&0\\ 0&0&0&9090\\ \end{matrix} \right) \end{equation*} For a fixed $j$, why these $j$-th invariant factors of $W_i$'s are all the same?

$\endgroup$

1 Answer 1

1
$\begingroup$

You can find the Smith normal form of $W_i$ using Gaussian Elimination. In this way, if you do this, let us say, to $W_4$, for instance, you start eliminating from the first columns, so you are doing a Gaussian elimination to $W_j$, $j<4$.

Please see Algebraically-nice general solution for last step of Gaussian elimination to Smith Normal Form? and Computing the Smith Normal Form on Math.StackExchange, and find more searching for "\(A\) Smith Normal Form" on SearchOnMath.

$\endgroup$
1
  • 1
    $\begingroup$ The string \(A\) is treated as bracketed by (uselessly) escaped parentheses: (A). I assumed you wanted the slashes to appear in the rendered output, in which case you have to escape them, too: \(A\) \\(A\\), and edited accordingly. I hope that this was correct. $\endgroup$
    – LSpice
    Apr 8, 2022 at 19:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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