Let $\|.\|$ be a norm on $\mathbb{R}^n$, and let $A$ be an $n\times n$ matrix. Put $\|x\|'=\|Ax\|$. What are the precise conditions on $A$ to ensure that $\|.\|'$ is also a norm?
I have to check the following:
I think that $(2)$ and $(3)$ can be fulfilled for any matrix because $\|\lambda x\|'=\|\lambda(Ax)\|=|\lambda|\|Ax\|=|\lambda|\|x\|'$ and $\|x+y\|'=\|A(x+y)\|=\|Ax+Ay\|\leq \|Ax\|+\|Ay\|=\|x\|'+\|y\|'$, I think the problem is in $(1)$ and I do not know what conditions $A$ must meet, could someone help me please? With being a diagonal matrix is it fulfilled? Thank you very much.

You need to know that $\lVert x \rVert' = \lVert Ax \rVert > 0$ for any $x \neq 0$. We know that $\lVert y \rVert > 0$ for any $y \neq 0$ because $\lVert \cdot \rVert$ is a norm, while $\lVert 0 \rVert = 0$. So you need that $Ax \neq 0$ for any $x \neq 0$. For $n \times n$ real matrices, this is the same as demanding that $A$ is invertible.