It is well known that a convergent sequence implies that its Cesàro mean converges while the converse is wrong. Then I am wondering the following:

  1. Would a bounded sequence imply the convergence of its Cesàro mean?
  2. Would the convergent Cesàro mean of a sequence imply that the sequence is bounded?

It would be appreciated if any proofs and examples are provided.


No and no.

  1. Consider the sequence $1,-1,1,1,-1,-1,1,1,1,1,-1,-1,-1,-1,1,1,1,1,1,1,1,1,-1,-1,-1,-1,-1,-1,-1,-1,\ldots $ where we have runs of $1$s or $-1$s with length $2^k$. It is a bounded sequence, but its Cesaro means are not convergent;

  2. Consider the sequence $\{a_n\}_{n\geq 1}$ where $a_n=1$ unless $n=2^{m}$ for some $m\in\mathbb{N}$, where $a_n=\log_2(n)=m$. The Cesaro means of $\{a_n\}_{n\geq 1}$ converge to $1$, but $\{a_n\}_{n\geq 1}$ is an unbounded sequence.


1. No. Let $\{a_n\}$ be defined as follows: $a_1=1$. $a_2=a_3=...=a_{n_1}=-1$, where $n_1>1$ is such that $|\frac{a_1+a_1+...+a_{n_1}}{n_1}-(-1)|<1/2$. Define $a_{n_1+1}=a_{n_1+2}=...=a_{n_2}=1$ where $a_{n_2}$ is such that $|\frac{a_1+a_1+...+a_{n_2}}{n_2}-1|<1/2$. And so on, inductively.

2. No. Let $a_n=\sqrt{n}$ when $n$ is a perfect square and $0$ otherwise. It is easy to see that $\frac{a_1+...+a_n}{n}\to0$


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.