Scribd
Upload
10 views3 pages

Homework 2

This document is a homework assignment for a measure theory course. It consists of 5 questions covering topics like σ-algebras, measures, completion of measures, and examples of probability measures. Students are instructed to submit a PDF file with their solutions by a certain deadline and to explain their work clearly. They are allowed to discuss the problems but must submit their own solutions.

Uploaded by

Muhammed Burak Eryılmaz
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views3 pages

Homework 2

This document is a homework assignment for a measure theory course. It consists of 5 questions covering topics like σ-algebras, measures, completion of measures, and examples of probability measures. Students are instructed to submit a PDF file with their solutions by a certain deadline and to explain their work clearly. They are allowed to discuss the problems but must submit their own solutions.

Uploaded by

Muhammed Burak Eryılmaz
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

IE 523 Homework 2 Fall 2023

Due: 3 October 2023 Tuesday, 17:30, through MOODLE.


Submit a single portable document file with a name formatted as follows:

<Lastname><Firstname>HW2.pdf

For instance, FellerWilliamHW2.pdf.

Write your solutions legibly and explain your mathematical arguments in full.
You are welcome to discuss the homework problems with your classmates but you
have to write down your own solutions.

The numbered references in the questions are from the lecture notes on MOODLE.

Q. 1. (10 pts) Solve Exercise 1.2.2.


Q. 2. (20 pts)

(a) (5) Let D be a countable subset of R. Show that B(D), which is defined as the
trace of B(R) on D, is equal to the discrete σ-algebra on D.
(b) (10) Solve Exercise 1.2.9.
Hint: Let T be the trace of F on A. The more involved part is to show σ(A) ⊇ T .
For this, defining the collection

D = {B ⊆ Ω | B ∩ A ∈ σ(A)} .

might be useful.
(c) (5) Solve Exercise 1.2.10.
N
Q. 3. (30 pts) Consider the infinite coin-tossing space Ω = {0, 1} . For each i ∈ N,
let Ai be the event that the ith toss yields heads, that is,

Ai = {ω = (ω1 , ω2 , . . .) ∈ Ω | ωi = 1} .

For each n ∈ N, define the σ-algebra

Fn = σ ({A1 , . . . , An }) .

(a) (5) Show that Fn ⊆ Fn+1 for each n ∈ N.


(b) (10) Let n ∈ N. Show that every event in Fn is a finite union of sets of the form
!  
\ \
BI = Ai ∩  Aci  ,
i∈I i∈{1,...,n}\I

where I ⊆ {1, . . . , n}.

1
S
(c) (5) Show that n∈N Fn is an algebra: it contains Ω as an element, it is closed
under complements and finite unions.
S
(d) (5) Show that n∈N Fn is not a σ-algebra.

One can consider the infinite coin-tossing game as a random experiment that evolves in
discrete time in the sense that the tosses are performed in consecutive days. Then, for
each n ∈ N, Fn represents the body of information that is accumulated during the first
n days. The “increasing” sequence (Fn )n∈N is called a filtration. As part (d) shows,
the union of all the members of this filtration does not produce a σ-algebra. For this
reason, one usually works with the σ-algebra
!
_ [
F∞ = Fn = σ Fn ,
n∈N n∈N

which denotes the information that is accumulated during the entire game.

(e) (5) Show that


F∞ = σ ({A1 , A2 , . . .}) .
Q. 4. (30 pts) Let (Ω, F, µ) be a measure space. Recall that a set A ∈ F is said to be
null if µ(A) = 0.

(a) (3) Show that a countable union of null sets is null.

Let us give a similar definition for an arbitrary subset: a set N ⊆ Ω (without any mea-
surability requirement) is said to be negligible if N is a subset of a null set. Note that
every null set is negligible; however, there can be negligible sets that are not measur-
able, in general.

(b) (3) Show that a countable union of negligible sets is negligible.

The first aim of this exercise is to construct a σ-algebra which is a minimal enlargement
of F that covers all negligible sets. To that end, let N be the collection of all negligible
subsets of Ω. Let F̄ be the σ-algebra generated by F and N , that is,
F̄ = σ(F ∪ N );
it is called the completion of F with respect to µ.

(c) (8) Show that


F̄ = {A ∪ N | A ∈ F, N ∈ N } .
Warning: Be careful about the notation: F ∪ N and {A ∪ N | A ∈ F, N ∈ N }
are not the same thing! The first one is a collection of subsets which precisely
consists of the sets in F and the sets in N . On the other hand, every set in the
second one is the union of a set in F and a set in N .
Hint: For the ⊆ part, first show that {A ∪ N | A ∈ F, N ∈ N } is a σ-algebra.

2
The next aim is to define a measure on F̄ which extends µ. Thanks to (c), let us define
a set function µ̄ : F̄ → [0, +∞] by

µ̄(F ) = µ(A)

for each F ∈ F̄ such that F = A ∪ N for some A ∈ F and N ∈ N ; µ̄ is called the


completion of µ.

(d) (8) Show that µ̄ is well-defined. In other words, show that if F ∈ F̄ such that
F = A ∪ N = B ∪ M for some A, B ∈ F and N, M ∈ N , then one must have
µ(A) = µ(B). This ensures that the definition of µ̄(F ) has no ambiguity.

(e) (5) Show that µ̄ is a measure on (Ω, F̄). Note that F ⊆ F̄. Show that the restriction
of µ̄ on F is µ, that is, if A ∈ F, then µ̄(A) = µ(A).
(f) (3) Show that (Ω, F̄, µ̄) is a complete measure space in the sense that every negli-
gible set of it is also measurable, that is, the negligible sets and the null sets of this
measure space are the same sets.
Q. 5. (10 pts)

(a) (5) Prove that a Dirac measure as defined in Example 1.3.8 is a probability measure.

(b) (5) Prove that a discrete measure as defined in Example 1.3.12 is a measure. Be
rigorous when changing the order of summations.

You might also like