10

Let a measure be $\mu:\Sigma\to\mathbb{R}_+$.

Call a measurable $A\subset\Sigma$ an atom if: $$\mu(A)>0:\quad\mu(E)<\mu(A)\implies\mu(E)=0\quad(E\subseteq A)$$ and a singleton $\{a\}\in\Sigma$ a point mass if: $$\mu(\{a\})>0$$

First of all, for singletons the notion of atoms and point masses coincide.

Next, for sigma-finite spaces the defintion of an atom is equivalent to: $$\mu(A)>0:\quad\mu(A\cap E)=0\lor\mu(A\cap E^c)=0$$ (For more details see: Measure Atoms: Definition?)

That is every atom splits into smaller atoms, so one could try to apply Zorn's lemma here.

Now, given a sigma-finite Borel measure $\lambda:\mathcal{B}(\Omega)\to\mathbb{R}_+$.

Does every atom arise from a point mass?

freishahiri
  • 17,045
  • 1
    Aren't you missing something akin to $E\subset A$, $E$ being a proper subset? – Alex R. Oct 23 '14 at 18:07
  • Yepp, added it... – freishahiri Oct 23 '14 at 18:11
  • By the way, your Zorn argument doesn't work. The natural thing to do is to partially order the atoms in $\Sigma$ by reverse inclusion. Given a chain ${E_i}_{i \in I}$ of atoms, to invoke Zorn you need to show it has an upper bound, i.e. an atom $E$ with $E \subset E_i$ for all $i$. The obvious candidate is $E := \bigcap_i E_i$, but if the chain is uncountable then this is an uncountable intersection, and you can't be sure that its measure is the same as that of the $E_i$s, or even that it is measurable at all. – Nate Eldredge Oct 23 '14 at 18:17
  • @NateEldredge: Yepp, thats a problem. But that would suggest to pick that point to construct a counterexample for a Borel algebra having atoms but no point masses, right? – freishahiri Oct 23 '14 at 18:20
  • I am not sure what you mean by that. – Nate Eldredge Oct 24 '14 at 00:48

2 Answers2

6

The example you give actually is a $\sigma$-finite Borel measure. Equip $[0,1]$ with the cofinite topology (in which a set is open iff it is either empty or its complement is finite). Then your $\Sigma$ is the Borel $\sigma$-algebra of the cofinite topology (it is a nice exercise to verify this).

However, there is the following result:

Proposition. Let $(X,d)$ be a separable metric space, $\Sigma$ its Borel $\sigma$-algebra, and $\mu$ a $\sigma$-finite measure on $\Sigma$. Then each atom of $\mu$ is the union of a point mass and a null set.

Proof. Let $C$ be a countable dense subset of $X$. For each integer $k\in\mathbb{N}$, we have $\bigcup_{x \in C} B(x, 1/k) = X$. Thus $\bigcup_{x \in C} (A \cap B(x,1/k)) = A$. So by countable additivity, there exists $x_k \in C$ such that $\mu(A \cap B(x_k, 1/k)) > 0$. Since $A$ is an atom, $\mu(A \setminus B(x_k, 1/k)) = 0$. Let $S = \bigcap_k B(x_k, 1/k)$.
Since for each $k$, $S$ is contained in a ball of radius $1/k$, $S$ contains at most one point.
On the other hand, by De Morgan's law and countable additivity, $$\mu(A \setminus S) = \mu\left(\bigcup_k A \setminus B(x_k, 1/k)\right) = 0.$$ Since $\mu(A\cap S)=\mu(A) > 0$, $A\cap S$ is not empty, so $A\cap S$ is a singleton.
Hence $A\cap S$ is a point mass and $A \setminus S$ a null set. $\Box$

So in this case, effectively the only atoms are point masses.

Note that we did not need to assume $X$ was complete.

For non-separable metric spaces, things are harder. For uncountable discrete spaces (which are certainly metric), the question of whether there can be nontrivial atoms is related to whether the cardinality of $X$ is a measurable cardinal, and such questions tend to be independent of the axioms of ZFC. I asked a new question about it: Consistency strength of 0-1 valued Borel measures.

Nate Eldredge
  • 101,664
  • So for non-separable metric spaces the question is not certainly answerable? – freishahiri Oct 24 '14 at 00:22
  • Don't you also need completeness besides searability? – freishahiri Oct 24 '14 at 00:34
  • In the proof I sketched in my head, I don't think I needed completeness, but I will have to check. Regarding non-separable spaces, I'm not completely sure. If there exists a measurable cardinal $\kappa$ then $\kappa$ with the discrete topology, equipped with a measure that assigns measure 1 to any set of cardinality $\kappa$ and measure 0 to everything else, is an atom that is not a point mass. But there may conceivably be another way to construct such a thing in ZFC. I just do not know. – Nate Eldredge Oct 24 '14 at 00:45
  • @Freeze_S: I added a proof of my proposition; we do not need completeness. – Nate Eldredge Oct 24 '14 at 05:42
  • Great! So everything works out fine =) – freishahiri Oct 24 '14 at 06:45
  • I linked your answer on a summary thread with mentioning you. Is this ok for you? http://math.stackexchange.com/questions/989196/borel-measures-atoms-summary – freishahiri Oct 24 '14 at 15:16
1

No, in general this is wrong!

Consider the Borel algebra generated by the cofinite topology: $$\Sigma:=\{E\subseteq[0,1]:\#E\leq\aleph_0\lor\#E^c\leq\aleph_0\}=\sigma(\mathcal{T}_{cofinite})$$ (See the thread on: Cofinite Topology: Borel Algebra)
and the finite Borel measure: $$\lambda(\#E\leq\aleph_0):=0$$ $$\lambda(\#E^c\leq\aleph_0):=1$$ Then the atoms are all the uncountables but no singleton is a point mass.

freishahiri
  • 17,045