Welcome to MathOverflow

Question

1

Dear all,

Is there any possible way to construct a set $A \subseteq \mathbb{R}^n $ for which $ H^{n-1} (\partial A) > Leb^ + (A ) $?

Where $ H^{n-1} (\partial A) $ is the Hausdorff measure of the boundary of $A$ and: $ Leb^{+} (A) = \lim_{\epsilon \to 0 } \frac{ Leb(A_ \epsilon) - Leb(A) }{\epsilon} $ , $A_\epsilon := { x \in \mathbb{R} ^n | d(x,A) \leq \epsilon } $ =Minkowski's content with respect to Lebesgue measure.

Thanks in advance !

edited Jul 25 at 8:31

asked Jul 12 at 18:34

Jason Mraz
201●6

Gerald Edgar · Accepted Answer · 2012-07-12 19:21:26Z UTC

3

So, $A$ is just a "set"? Say $A$ is an open square minus a mid-line: $$ A = \{ (x,y): 0 \lt x \lt 1 \text{ and (} 0 \lt y \lt 1/2 \text{ or } 1/2 \lt y \lt 1\text{)}\} $$ Then $\partial A$ consists of 5 line-segments of length $1$ (the 4 sides and the mid-line), so $H^1(\partial A) = 5$. But $\mathrm{Leb}^+(A) = 4$, missing the mid-line.

answered Jul 12 at 19:21

Gerald Edgar
15.4k●1●29●67

Thanks a lot ! The problem is that I can't understand the calculation of such thing: By definition : $H^1 ( \partial A) = lim _{\delta \to 0} inf \{ \sum C diam(U_i) | \partial A \subseteq \cup U_i , diam(U_i ) \leq \delta \} $ Can you please detail a bit about how you deduced that $H^1 (\partial A) = 5$ in this case? Thanks a lot @! – Jason Mraz Jul 12 at 20:18

2

First: For a line segment of length $a$, the one-dimensional Hausdorff measure is $a$. Hausdorff measure is additive on Borel sets, so the measure for the 5-segment set is $5$. – Gerald Edgar Jul 12 at 21:13

Different Measures On R2

Remember to vote up questions/answers you find interesting or helpful (requires 15 reputation points)

1 Answer

You can accept an answer to one of your own questions by clicking the check mark next to it. This awards 15 reputation points to the person who answered and 2 reputation points to you.