1
$\begingroup$

On the Appendix C about Hausdorff dimensions and measures of the book Multiparamerer Processes by Khoshnevisan it is written the following

Lemma 1.1.2 For all Borel sets $E\subset R^d$, and for all real numbers $s\geq 0$, $$H_s(E)\geq Leb_s(E). $$

Proof Let us fix an arbitrary $\varepsilon>0$ and find arbitrarily closed $l^\infty$ balls $B_1, B_2...$ whose radii $r_1, r_2...$ are all less or equal to $\varepsilon$. Whenever $E\subset\cup_i B_i$ $$Leb_s(E)\leq\sum_i Leb_s(B_i)=\sum_i(2 r_i)^s $$ (...)

Where $H_s$ is of course the s-dimensional Hausdorff measure on $R^d$.

My question is the following: Do you know which is the definition of $Leb_s$? He refers to it as the $s$-dimensional Lebesgue measure on $R^d$.

I just know the standard definition of Lebesgue measure https://proofwiki.org/wiki/Definition:Lebesgue_Measure

I also checked on the rest of the book to see if there was a more concrete definition of $Leb_s$ and I didn´t find it... probably I missed it!

$\endgroup$
  • 1
    $\begingroup$ I've never seen any such definition, except of course for $s=d$. Possibly if you look at the proof you can figure out what measure he has in mind... $\endgroup$ – David C. Ullrich Apr 10 '18 at 13:32
  • $\begingroup$ From the proof you can deduce that if you take an $l_{\infty}$ ball $B$ of radii $r$ then $Leb_s(B)=(2r)^s$. $\endgroup$ – Adrián Hinojosa Calleja Apr 11 '18 at 11:05
  • $\begingroup$ Really? It's not hard to show that there is no such measure if $s\ne d$. $\endgroup$ – David C. Ullrich Apr 11 '18 at 12:07
  • $\begingroup$ Yes it's quite strange this doesn´t make any sense... I doble checked in the book and there is no reference in about what is $Leb_s$, I thought it could be a typo and it should say $Leb^s$ but even in this case it doesn´t make any sense... I wrote part of the proof in the post. $\endgroup$ – Adrián Hinojosa Calleja Apr 12 '18 at 20:15
1
$\begingroup$

I finally contacted the author of the book and there is a typo. The correct statement is

Lemma 1.1.2 For all Borel sets $E\subset \mathbb{R}^d$, and for all real numbers $s\geq d$, $$H_s(E)\geq Leb_d(E). $$

$\endgroup$

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.