Homepage › Solution manuals › Terence Tao › An Introduction to Measure Theory › Exercise 0.0.2 (Tonelli’s theorem for series over arbitrary sets)

Exercise 0.0.2 (Tonelli’s theorem for series over arbitrary sets)

Let A , B be sets (possibly infinite or uncountable), and ( x n , m ) n ∈ A , m ∈ B be a doubly infinite sequence of extended non-negative reals x n , m ∈ [ 0 , + ∞ ] indexed by A and B . Show that

∑ ( n , m ) ∈ A × B x n , m = ∑ n ∈ A ∑ m ∈ B x n , m = ∑ m ∈ B ∑ n ∈ A x n , m .

(Hint: although not strictly necessary, you may find it convenient to first establish the fact that if ∑ ⁡ n ∈ A x n is finite, then x n is non-zero for at most countably many n .)

Answers

We can consider two cases:

1.
the uncountable sum is infinite: the equation holds obviously.
2.
the sum is finite: recall last exercise 0.0.1, we just need to prove the conclusion holds when there are countable elements that are non-zero which has been proved in Exercise 0.0.1.
User profile picture
2024-08-19 11:47
Comments
style=”width:0%”

Proof of Tonelli’s Theorem for Series

Let x n , m ≥ 0 for all ( n , m ) ∈ A × B . We aim to prove:

∑ ( n , m ) ∈ A × B x n , m = ∑ n ∈ A ∑ m ∈ B x n , m
(1)

Let  F  be a finite set such that  F ⊂ A × B . (2) Define  A 1 = { x ∣ ∃ ⁡ y ∈ B , ( x , y ) ∈ F } , A 2 = { y ∣ ∃ ⁡ x ∈ A , ( x , y ) ∈ F } . (3) Then  A 1 , A 2  are finite sets,  A 1 ⊂ A , A 2 ⊂ B ,  and  F ⊂ A 1 × A 2 . (4) ∑ ( n , m ) ∈ F x n , m ≤ ∑ ( n , m ) ∈ A 1 × A 2 x n , m = ∑ n ∈ A 1 ∑ m ∈ A 2 x n , m ≤ ∑ n ∈ A ∑ m ∈ B x n , m . (5) Therefore,  ∑ ( n , m ) ∈ A × B x n , m = sup ⁡ F ⊂ A × B F  finite ∑ ( n , m ) ∈ F x n , m ≤ ∑ n ∈ A ∑ m ∈ B x n , m . (6) If  ∑ ( n , m ) ∈ A × B x n , m = + ∞ ,  then  ∑ n ∈ A ∑ m ∈ B x n , m ≤ ∑ ( n , m ) ∈ A × B x n , m . (7) If  ∑ ( n , m ) ∈ A × B x n , m < + ∞ : (8) Let  C = { ( n , m ) ∈ A × B ∣ x n , m ≠ 0 } . (9) A 3 = { x ∣ ∃ ⁡ y ∈ B , ( x , y ) ∈ C } , (10) B n = { m ∈ B ∣ ( n , m ) ∈ C } . (11) Then  A 3 ⊂ A , B n ⊂ B ,  and  C ⊂ A × B  are all at most countable sets . (12) Hence,  ∑ n ∈ A ∑ m ∈ B x n , m = ∑ n ∈ A 3 ∑ m ∈ B n x n , m = ∑ ( n , m ) ∈ C x n , m ≤ ∑ ( n , m ) ∈ A × B x n , m . (13)

Conclusion

Combining the two cases, we have shown that:

∑ ( n , m ) ∈ A × B x n , m = ∑ n ∈ A ∑ m ∈ B x n , m .
(14)
User profile picture
2025-10-30 13:00
Comments