Exercise 8.2.3

Assume that μ has properties 0)-ii). Prove properties iv) and v). Also prove:

vi)
μ ( A B ) = μ ( A ) + μ ( B ) μ ( A B ) .
vii)
μ ( n = 0 A n ) n = 0 μ ( A n ) .

Answers

First, for reference, we assume the following properties of μ :

0)
μ ( [ a , b ] ) = b a for any a and b in 𝑹 where a < b .
i)
μ ( ) = 0 and μ ( 𝑹 ) = .
ii)
If { A n } n = 0 is a collection of mutually disjoint subsets of 𝑹 , then μ ( n = 0 A n ) = n = 0 μ ( A n ) .

First we show

iv)
If A B = then μ ( A B ) = μ ( A ) + μ ( B ) .

Proof. Assume that A B = and define A 0 = A , A 1 = B , and A n = for all natural n > 1 . Then clearly each of the sets in { A n } n = 0 are mutually disjoint. It is also trivial to show that n = 0 A n = A B . We then have by property ii) that

μ ( A B ) = μ ( n = 0 A n ) = n = 0 μ ( A n ) = μ ( A 0 ) + μ ( A 1 ) + n = 2 μ ( A n ) = μ ( A ) + μ ( B ) + n = 2 μ ( ) = μ ( A ) + μ ( B ) + n = 2 0 = μ ( A ) + μ ( B ) ,

noting that we have also used property i) according to which μ ( ) = 0 . This shows the desired result. □

Lemma 1. For a measure μ , if A B 𝑹 then μ ( B A ) = μ ( B ) μ ( A ) .

Proof. Clearly A and B A are disjoint sets such that A ( B A ) = B so that μ ( A ) + μ ( B A ) = μ ( B ) by property iv). The result then clearly follows by subtracting μ ( A ) from both sides. □

Next we show

v)
If A B then μ ( A ) μ ( B ) .

Proof. Suppose that A B so that μ ( A ) + μ ( B A ) = μ ( B ) by Lemma 1. Since μ is a function into [ 0 , ) { } , it follows that

0 μ ( B A ) μ ( A ) μ ( B A ) + μ ( A ) = μ ( A ) + μ ( B A ) μ ( A ) μ ( B )

as desired. □

Now we show

vi)
μ ( A B ) = μ ( A ) + μ ( B ) μ ( A B ) .

Proof. Let C = A B , A = A C , and B = B C . It is then trivial to show that A , B , and C are mutually disjoint sets such that A B C = A B . We then have by a straightforward extension of property iv) that

μ ( A ) + μ ( B ) + μ ( C ) = μ ( A B ) μ ( A C ) + μ ( B C ) + μ ( C ) = μ ( A B ) μ ( A ) μ ( C ) + μ ( B ) μ ( C ) + μ ( C ) = μ ( A B ) μ ( A ) + μ ( B ) μ ( C ) = μ ( A B ) μ ( A ) + μ ( B ) μ ( A B ) = μ ( A B )

as desired. Note that we have also used Lemma 1 since clearly C A and C B so that μ ( A C ) = μ ( A ) μ ( C ) and μ ( B C ) = μ ( B ) μ ( C ) . □

Lastly we show

vii)
μ ( n = 0 A n ) n = 0 μ ( A n ) .

Proof. Supposing that we have a system of sets { A n } n = 0 , first we define a sequence of corresponding sets recursively:

A 0 = A 0 A n + 1 = A n + 1 k = 0 n A k

Note that it is clear that A n A n for any n 𝑵 so that μ ( A n ) μ ( A n ) by property v).

We now show that each of these sets are mutually disjoint. So consider any natural m and n where m n . Without loss of generality, we can then assume that m < n . Suppose that A m and A n are not disjoint so that there is an x A m A n . Thus x A n = A n k = 0 n 1 A k so that x k = 0 n 1 A k . However since also x A m and 0 m n 1 , we also can conclude that x k = 0 n 1 A k . Since this is a contradiction, it must be that A m and A n are in fact disjoint, which shows mutual disjointedness since m and n were arbitrary.

Next we show that n = 0 A n = n = 0 A n . The direction is clear since, for any x n = 0 A n , there a natural n where x A n . Since A n A n it follows that x A n so that clearly x n = 0 A n . Now consider any x n = 0 A n so that there is a natural n where x A n . Clearly if n = 0 then x A 0 = A 0 . So assume that n > 0 so that k = 0 n 1 A k is defined. If x k = 0 n 1 A k then there is a 0 k n 1 where x A k . On the other hand, if x k = 0 n 1 A k then clearly x A n k = 0 n 1 A k = A n . Thus in all cases there is a natural k such that x A k so that x n = 0 A n as desired.

We therefore have

μ ( n = 0 A n ) = μ ( n = 0 A n ) = n = 0 μ ( A n ) (by property ii) since the sets  { A n } n = 0  are mutually disjoint) n = 0 μ ( A n ) (since  0 μ ( A n ) μ ( A n )  for all natural  n )

as desired. □

User profile picture
2024-07-15 11:42
Comments