My Math Forum A limit

 Calculus Calculus Math Forum

 August 27th, 2019, 07:27 AM #1 Senior Member   Joined: Dec 2015 From: somewhere Posts: 642 Thanks: 91 A limit Evaluate $\displaystyle \lim_{n\rightarrow \infty }n^{-1} \underbrace{\ln \ln...\ln}_{n}(n)$. Last edited by skipjack; August 27th, 2019 at 07:54 AM.
 August 27th, 2019, 07:53 AM #2 Global Moderator   Joined: Dec 2006 Posts: 20,972 Thanks: 2222 Did you invent this problem?
 August 27th, 2019, 12:26 PM #3 Math Team   Joined: Dec 2013 From: Colombia Posts: 7,683 Thanks: 2664 Math Focus: Mainly analysis and algebra Mmm... since $\ln x \to -\infty$ as $x \to 0$ at an exponential rate, I'm going to guess that this blows up. But I'm far from certain.
August 27th, 2019, 12:35 PM   #4
Senior Member

Joined: Aug 2012

Posts: 2,393
Thanks: 749

Quote:
 Originally Posted by v8archie Mmm... since $\ln x \to -\infty$ as $x \to 0$ at an exponential rate, I'm going to guess that this blows up. But I'm far from certain.
The limit is to infinity. $\log x$ goes to infinity as does any number of iterations of $\log$. It's an old joke that $\log \log x$ is constant for all practical purposes, the point being that it goes to infinity very slowly. But to infinity it goes nonetheless. So I don't follow your intuition. The intuition would have to be that the iterated log goes to infinity so slowly that the n in the denominator pulls it to zero.

It looks like you could prove it by using L'Hospital perhaps.

August 27th, 2019, 01:25 PM   #5
Senior Member

Joined: Jun 2019
From: USA

Posts: 213
Thanks: 90

*sigh* Mathematicians...

It sure looks like it's going towards zero to me.
Attached Images
 lnnn.jpg (13.9 KB, 4 views)

August 27th, 2019, 02:37 PM   #6
Senior Member

Joined: Jun 2019
From: USA

Posts: 213
Thanks: 90

In fact, the magnitude is shrinking as 1/n.
Attached Images
 lnnnloglog.jpg (18.0 KB, 3 views)

 August 27th, 2019, 03:32 PM #7 Senior Member   Joined: Sep 2016 From: USA Posts: 645 Thanks: 408 Math Focus: Dynamical systems, analytic function theory, numerics The limit is 0. You can easily see this by noticing that $\log^{(k)}(n) \leq \log n$ for all $k \geq 1$ where the notation on the left means iteration. Combine this with the fact that $n^{-1} \log n \to 0$.
August 27th, 2019, 03:49 PM   #8
Senior Member

Joined: Jun 2019
From: USA

Posts: 213
Thanks: 90

Quote:
 Originally Posted by SDK $\log^{(k)}(n) \leq \log n$ for all $k \geq 1$
I thought of this, but the iterated logs quickly become negative (and thus complex). If $| \log^{(k)}(n) | \leq |\log n|$, then it still works, but I wasn't sure how to determine that. Plotting was easier.

 Tags limit

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post zylo Calculus 13 May 31st, 2017 12:53 PM dunn Calculus 1 November 6th, 2011 08:12 AM veronicak5678 Real Analysis 4 August 22nd, 2011 10:07 AM mathman2 Calculus 1 October 22nd, 2009 06:44 PM conjecture Calculus 1 July 24th, 2008 01:14 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top