MA 408: Measure Theory
Practice Set – 1 (Solutions)
Question 1
Let X be any nonempty set. Define
C := {E ⊆ X : either E or E c is finite}.
Prove that C is an algebra on X. Show that if X is finite, then C = P(X).
Solution
To prove that C is an algebra, we verify:
(i) X ∈ C
Since X c = ∅ is finite, X ∈ C.
(ii) Closed under complement
If E ∈ C, then either E or E c is finite.
If E is finite, then E c is cofinite. If E c is finite, then E is cofinite.
Thus in either case E c ∈ C.
(iii) Closed under finite unions
Let E, F ∈ C.
Case 1: Both finite ⇒ E ∪ F finite.
Case 2: Both cofinite. Then
(E ∪ F )c = E c ∩ F c
which is intersection of finite sets, hence finite.
Case 3: One finite, one cofinite. Union is cofinite.
Hence C is an algebra.
If X is finite, then every subset is finite. Thus C = P(X).
1
Question 2
Show that intersection of two semi-algebras need not be a semi-algebra.
Solution
Let X = R.
Define
S1 = {(a, b] : a < b} ∪ {∅}
S2 = {[a, b) : a < b} ∪ {∅}
Then
S1 ∩ S2 = {∅}
A semi-algebra must contain X. But X ∈
/ {∅}.
Hence not a semi-algebra.
Question 3
Let F and G be semi-algebras on X and Y respectively. Show that
F × G = {F × G : F ∈ F, G ∈ G}
is a semi-algebra on X × Y .
Solution
(i) Contains X × Y
Since X ∈ F and Y ∈ G,
X × Y ∈ F × G.
(ii) Closed under intersection
(F1 × G1 ) ∩ (F2 × G2 ) = (F1 ∩ F2 ) × (G1 ∩ G2 )
which belongs to F × G.
(iii) Complement property
(F × G)c = (F c × Y ) ∪ (F × Gc ).
Since complements in semi-algebra decompose into finite disjoint unions, the result is
finite union of rectangles.
Hence semi-algebra.
2
Question 4
Let C be a semi-algebra on X and ∅ ̸= E ⊆ X. Define
C ∩ E = {A ∩ E : A ∈ C}.
Show it is a semi-algebra on E.
Solution
(i) Contains E
E = X ∩ E and X ∈ C.
(ii) Intersection closed
(A ∩ E) ∩ (B ∩ E) = (A ∩ B) ∩ E.
(iii) Complement in E
E \ (A ∩ E) = (X \ A) ∩ E.
Thus semi-algebra.
Question 5
Let f : X → Y and C semi-algebra on Y . Show that
f −1 (C) = {f −1 (E) : E ∈ C}
is semi-algebra on X.
Solution
Using properties of inverse image:
f −1 (Y ) = X
f −1 (E1 ∩ E2 ) = f −1 (E1 ) ∩ f −1 (E2 )
f −1 (E c ) = (f −1 (E))c
Thus semi-algebra.
3
Question 6
Let F , G be algebras. Is F × G an algebra?
Solution
In general, no.
Complement:
(A × B)c = (Ac × Y ) ∪ (A × B c )
This is union of rectangles, not a single rectangle.
Thus not closed under complement.
Hence not algebra in general.
Question 7
If two finite measures agree on generating set S, do they agree on F(S)?
Solution
If S is a π-system and measures finite, then by uniqueness theorem (Dynkin’s π-λ theorem),
they agree on σ(S).
Otherwise not necessarily.
Question 8
Show:
µ(E ∪ F ) + µ(E ∩ F ) = µ(E) + µ(F ).
Solution
Write
E ∪ F = E ∪ (F \ E).
Using additivity:
µ(E ∪ F ) = µ(E) + µ(F ) − µ(E ∩ F ).
Rearranging gives result.
Question 9
Show each given collection generates B(R).
4
Solution
Open intervals generate topology.
Example:
∞
[ 1 1
(a, b) = [a + , b − ].
n=1
n n
Similarly:
∞
[
(a, +∞) = (a, n)
n=1
Thus all collections generate open sets, hence Borel σ-algebra.
Question 10
Define Dirac measure: (
1 a∈E
δa (E) =
0 a∈
/ E.
Solution
(i) Null empty set
δa (∅) = 0.
(ii) Countable additivity
In disjoint family, at most one set contains a. Thus additivity holds.
Hence measure.
δa (R) = 1.
Question 11
Let E1 ∆E2 be symmetric difference.
Solution
(a) If µ(E1 ∆E2 ) = 0, then
E1 = (E1 ∩ E2 ) ∪ (E1 \ E2 ).
Since difference is null, measures equal.
(b) Completeness implies subsets of null sets measurable.
5
Question 12
Properties of Lebesgue measure.
Solution
(a) By definition of outer measure.
(b) For singleton {x}, cover by interval of length ϵ.
Thus measure 0.
(c) Countable union of singletons → 0.
(d) Translation invariance follows from length invariance.
(e) Dilation scales length:
λ(tA) = |t|λ(A).
Question 13
Verify probability measures.
Solution
(a) Bernoulli:
1
P ({H}) = P ({T }) = , P (Ω) = 1.
2
(b) Binomial:
N
X N k
p (1 − p)N −k = (p + 1 − p)N = 1.
k=0
k
(c) Poisson:
∞
X λk
e−λ = e−λ eλ = 1.
k=0
k!
Odd integers probability:
1 − e−2λ
P (odd) = .
2
Prime integers:
X λp
P (primes) = e−λ .
p∈P
p!