Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $\rho(A)$ be the spectral radius of $A,$ that is the maximal eigenvalue of $A$ in absolute value. I want to show that for any $ \rho(A) < \eta < 1,$ there is $c>0$ such that $\|A^k\| \leq c \eta^k$ for $k = 0,1, \dots.$ Can somebody show me how to do this? Thanks!

share|cite|improve this question
You'll have to be more specific about what you mean by $\|A^t\|$. Which norm? Spectral norm, Frobenius norm, ...? – John Moeller Dec 17 '12 at 22:27
Any matrix norm! could this work with any matrix norm? aren't that norms are equivalent in finite dimensional space? – Zizo Dec 17 '12 at 22:41
What is $\eta^t$ here? And do you mean "... want to show that there is $c>0$ so that for any $\eta>\rho(A)$ ... " ? That is, your quantifiers appear to be in the wrong order, and I don't see why $\eta<1$ might be important. – user108903 Dec 17 '12 at 22:45
$\eta=100000000$ trivially satisfies the problem with $c=1$ otherwise. – Inquest Dec 17 '12 at 22:45
@user108903 it does not matter, we could show that $ \|A^t\| < \rho(A)^t$ instead! Because this would implies that $\|A^t\| < \rho(A)^t.$ Since there is $c>0$ such that $ \rho(A)^t \leq c \eta^t$ for $t = 1,2, \dots$ then the inequality would follow. So the question would be, is it true that $\|A^t\| < \rho(A)^t$ for a matrix norm. I want the proof to work for any arbitrary matrix norm – Zizo Dec 17 '12 at 22:54
up vote 3 down vote accepted

Let $B=\eta^{-1}A$. Then $\rho(B)<1$. Hence $B^k\rightarrow0$ as $k\rightarrow\infty$. Therefore the entries of $B^k$ are bounded. Hence the Frobenius norm of $B^k$ is bounded and in turn, $\|B^k\|$ is also bounded because all matrix norms are equivalent. Therefore $\|B^k\|\le c$ for some $c>0$, i.e. $\|A^k\|\le c\eta^k$.

share|cite|improve this answer
That's what I am talking about.. Nice proof. – Zizo Dec 18 '12 at 6:12

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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