Given a sequence of complex numbers $\{a_n\}_n$, one says that this sequence admits $a$ as a sequential density if $$\underset{N_s\to\infty}{\lim}\frac{1}{N_s}\sum_{n=1}^{N_s} a_n = a$$ where $N_s = 2^{2^s}$ for instance. The sequence admits $a$ as a logarithmic density if $$\underset{N\to\infty}{\lim}\frac{1}{\log N}\sum_{n=1}^N \frac{1}{n} a_n = a.$$ If the the sequence has a sequential density, does it also have a logarithmic one, with the same limit ?

Motivation: In a paper called Problems of Almost Everywhere convergence related to harmonic analysis and number theory, Jean Bourgain states that for any function $f$ in $L^2(\mathbf{T})$ the sequence of its Riemann sums $\{R_nf(x)\}$ admits $\int_0^1f$ as a logarithmic density. In the proof, it turns out that Bourgain reduces this convergence to an $L^2$-maximal inequality

$$\left\Vert\underset{N_s=2^{2^s}}{\sup}\frac{1}{N}\left\vert\sum_{n=1}^NR_nf\right\vert\right\Vert_2\leq C \left\Vert f \right\Vert_2 $$

which actually proves the sequential density of the sequence $\{R_nf(x)\}$ for almost every $x$ in $\mathbf{T}$.

In number theory, the Davenport-Erdös theorem states the equivalence of this two notions of densities for sets. See this post for instance. I don't know where to find a proof of this theorem in order to adapt it for sequences.

Thank you in advance

  • $\begingroup$ Also posted on math.stackexchange $\endgroup$ – Wulfenite Aug 10 at 14:30
  • $\begingroup$ So it is considered impolite to simultaneously post on both sites. If you're not sure, try MSE first, and if there is no answer in 2 or 3 days, consider posting on MO. The issue is that this leads to wasted effort. $\endgroup$ – Anthony Quas Aug 10 at 16:17
  • 2
    $\begingroup$ So I think the answer to your question is "no" unless you have some kind of growth conditions on the $a_n$. For instance if $a_{2^{2^n}}=2^{2^n}$ and $a_{2^{2^n}+1}=-2^{2^n}$ and all other terms are 0, the first sequence has limit 1, whereas the second sequence has limit 0. $\endgroup$ – Anthony Quas Aug 10 at 16:30
  • $\begingroup$ As usually I will link to the discussion of cross-posting on MathOverflow Meta. (And a similar post on Meta Mathematics Stack Exchange. This is meant as a kind of follow up on Anthony Quas' comment.) $\endgroup$ – Martin Sleziak Aug 10 at 16:39
  • $\begingroup$ Sorry for publishing on the two forums. I understand and won't do that again, I'm not familiar with all of this. I was about to delate my post from MathStackExange when @Martin Sleziak answered me. Thanks a lot for your help. $\endgroup$ – Wulfenite Aug 10 at 22:19

The answer is "no" even if we assume that the $a_n$ are bounded. For example, take $a_n=1$ if the fractional part of $\log_2\log_2 n$ is between $0$ and $\frac12$ (equivalently, if $n$ is between $2^{2^k}$ and $2^{2^{k+1/2}}$ for some integer $k$), and $a_n=0$ otherwise. Then the sequential limit will equal $0$, but the lim sup of the logarithmic-density expression equals $1$ while its lim inf equals $0$.


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.