2
$\begingroup$

It seems that the matrix $K\times K$ matrix $\boldsymbol M$ with components \begin{equation} M_{nj}=(n+1)^j \end{equation} is invertible for any dimension $K>0$.

I don't konw how to show this, but it seems natural. My first approach would be to show that the determinant is nonvanishing, i.e. \begin{equation} \det M=\sum_{\rho\in S_K}\text{sgn}({\rho})\prod^K_{n=1}(n+1)^{\rho(n)} \end{equation} or its transpose: \begin{equation} \det M^T=\sum_{\rho\in S_K}\text{sgn}({\rho})\prod^K_{n=1}(\rho(n)+1)^{n} \end{equation} but I don't see how to go from there. But I think it is already interesting in its own right.

The motivation stems from the following question: Is it true that for any sequence of real numbers $\zeta_1,\zeta_2,\dots, \zeta_K$ I can find a sequence of real numbers $\dots,s_{-3},s_{-2},s_{-1},s_{0},s_{1}s_{2},\dots$ such that $\zeta_k=\sum_{n\in \mathbb Z} s_{n}n^k$ for all $k$. If the above matrix is invertible, then this is the case since $\vec \zeta=\boldsymbol M^T \vec s $ has a solution (here we just need $s_1,\dots, s_K$, so the other entries could be chosen as 0 to get the infinite sequence).

However this is just the motivation and the initial question is more interesting.

  • 1
    This matrix is invertible, since it is can be attained from a [Vandermonde matrix](https://en.wikipedia.org/wiki/Vandermonde_matrix)2017-01-13

1 Answers 1

2

$$ |M_{nj}|= \begin{vmatrix} 2^1 & 2^2 & 2^3 &...&2^n \\ 3^1 & 3^2 & 3^3 &...&3^n \\ ...\\ (n+1)^1 & (n+1)^2 & (n+1)^3 &...&(n+1)^n \\ \end{vmatrix}= $$

$$=2\cdot3\cdot4...\cdot(n+1) \begin{vmatrix} 1 & 2^1 & 2^2 &...&2^{n-1} \\ 1 & 3^1 & 3^3 &...&3^{n-1} \\ ...\\ 1 & (n+1)^1 & (n+1)^2 &...&(n+1)^{n-1} \\ \end{vmatrix}$$

So the last one is a Vandermonde Matrix wich is given by:

$$(n+1)!\cdot \prod_{1\le i\le j}(M_{i2}-M_{j2}) \ne0$$

once $M_{i2}\ne M_{j2}$.

  • 0
    Thanks Arnaldo for your answer. It's great, thanks. I learned a lot from this question. Didn't know about the Vandermonde matrix. On a first reading I always read Voldemort matrix although the words barely look similar.2017-01-14
  • 0
    You are very welcome.2017-01-14
  • 0
    sure. Was just about to..2017-01-14
  • 0
    MSE is in fact an amazing place for learn about math. Days ago I've learnt about gershgorin disk theorem. It was a silver bullet for a tough problem.2017-01-14