What is an upper-bound on $x$, given that $x < k \ln(1+x)$? I believe that the solution is something of the form $\mathcal{O}(k \ln k)$ but I am unable to prove this.

This is my first encounter with such an inequality and any pointers to relevant techniques will be helpful. Thanks.

  • $\begingroup$ Are you interested in the case when $k \to \infty$? $\endgroup$ – Antonio Vargas Feb 15 '15 at 16:58
  • $\begingroup$ I am not certain about it. Since in the cases I saw, $k$ is of the form $\frac{1}{\epsilon}$ (where $\epsilon$ is a sufficiently small error quantity) I believe what you say might be true. Can you provide a solution for that case? $\endgroup$ – Enigman Feb 15 '15 at 17:13

For $k>1$ let $X(k)$ be the unique positive solution of $x=l\log(1+x)$. Then $$ k\,(\log k+\log\log k)<X(k)<k\,\Bigl(\log k+\log\log k+\frac{1}{\sqrt{\log k}}\Bigr). $$ I will prove the first inequality. We want to prove that $$ k\,(\log k+\log\log k)<k\log(1+k\,(\log k+\log\log k)). $$ This is equivalent to $$ \log k+\log\log k<\log(1+k\,(\log k+\log\log k)), $$ which in turn is equivalent to the trivial inequality $$ k\log k<1+k\,(\log k+\log\log k). $$ The other inequality can be proved in a similar way, with a little more work.


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.