Proving Properties of Countable Sets & Probability Spaces

Click For Summary

Homework Help Overview

The discussion revolves around proving properties of countable sets and probability spaces, specifically addressing cardinality and measure theory. The original poster presents two problems: one concerning the existence of a countable family of sets in the power set of X and the other regarding the intersection of sets in a probability space.

Discussion Character

  • Mixed

Approaches and Questions Raised

  • Participants explore implications of cardinality and the structure of sigma algebras, questioning the sufficiency of certain set constructions. They discuss counting methods in the context of measure theory and consider the implications of measures associated with sets in a probability space.

Discussion Status

Some participants have offered guidance on potential approaches, such as applying combinatorial methods to measure theory. There is ongoing exploration of the implications of the sums related to the sets in the probability space, with no explicit consensus reached on the sufficiency of certain arguments.

Contextual Notes

Participants are navigating the complexities of measure theory and cardinality, with some constraints noted regarding the definitions and properties of the sets involved. There is also mention of specific properties of the probability space that may influence the discussion.

hellbike
Messages
61
Reaction score
0
1.prove that for any set X: |X|<c <=> in P(X) exist such countable set family F, that sigma algebra generated by F contains all points.

2.let (X,E,u) be probability space and A_1,...,A_2009 in E have property u(A_i)>=1/2. Prove that there exist x such is in A_i for atleast 1005 different i.

i need some tip in solving this.
 
Last edited:
Physics news on Phys.org
On problem 1, for the forward implication, note that if |X| ≤ c, then you can assume WLOG that X⊆R. For the reverse, show that the cardinality of the Σ-algebra generated by any countable family of sets is at most the continuum.

For problem 2, try applying a counting method. Specifically, define S_n as the set of all points that are in exactly n of the A_i. What is ∑nμ(S_n)?
 
What do you mean by counting method? Induction?

And this sum is between 1/2 and 1, but i don't know why this would be useful?

and for problem 1:
->
is this sufficient to say that set of (p,q) for p,q in Q can do the job for any subset of R?
 
Last edited:
hellbike said:
What do you mean by counting method? Induction?

I mean extending combinatorial methods to the measure-theoretic setting. In this case, we want to "count" (i.e. find the measure of) the set of points in the A_i with multiplicity (i.e. a point that appears in two different A_i would be counted twice) in two different ways. First, we divide it up into the disjoint sets S_n, and observe that each point in S_n will be counted exactly n times, so the total measure (with multiplicity) is \sum_{n=1}^{2009}n\mu (S_n). On the other hand, each point is counted once each time it appears in an A_i, so the total measure (with multiplicity) is \sum_{i=1}^{2009}\mu (A_i). So these two sums are equal. Can you see how this implies that μ(S_n)>0 for at least one n≥1005?

is this sufficient to say that set of (p,q) for p,q in Q can do the job for any subset of R?

Nearly, although since you're generating a Σ-algebra on X, you should write (p,\ q) \cap X,\ p,\ q\in\mathbb{Q}
 
yes, thank you very much.
 

Similar threads

Replies
1
Views
2K
Replies
20
Views
4K
  • · Replies 5 ·
Replies
5
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 12 ·
Replies
12
Views
2K
  • · Replies 31 ·
2
Replies
31
Views
5K
Replies
20
Views
5K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K