Scribd
Upload
5 views11 pages

Mat2125 Lec 01

The document contains solutions to exercises from a course on Elementary Real Analysis, specifically addressing properties of exponential functions and the set of positive real numbers. It demonstrates that for a real number c > 0, the behavior of cn depends on whether c is greater than or less than 1, and explores the existence of lower and upper bounds for the set S2 of positive real numbers. The proofs provided utilize concepts such as Bernoulli's Inequality and the Completeness Property of real numbers.

Uploaded by

sqzbmf6g9g
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
5 views11 pages

Mat2125 Lec 01

The document contains solutions to exercises from a course on Elementary Real Analysis, specifically addressing properties of exponential functions and the set of positive real numbers. It demonstrates that for a real number c > 0, the behavior of cn depends on whether c is greater than or less than 1, and explores the existence of lower and upper bounds for the set S2 of positive real numbers. The proofs provided utilize concepts such as Bernoulli's Inequality and the Completeness Property of real numbers.

Uploaded by

sqzbmf6g9g
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/ 11

MAT 2125

Elementary Real Analysis

Exercises – Solutions – Q2-Q4

Winter 2021

P. Boily, A. Smith (uOttawa)


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

2. Let c > 0 be a real number.


(a) If c > 1, show that cn ≥ c for all n ∈ N and that cn > 1 if n > 1.
(b) If 0 < c < 1, show that cn ≤ c for all n ∈ N and that cn < 1 if n > 1.

P. Boily, A. Smith (uOttawa) 1


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

Proof. The statement is clearly not true if n = 0: as a result, we


must interpret N to stand for the set N = {1, 2, 3, . . .}, without the 0.
Generally, we use whatever “version” of N is appropriate.
(a) If c > 1, ∃x ∈ R such that x > 0 and c = 1 + x. Let n ∈ N. First
note that n − 1 ≥ 0 and so (n − 1)x > 0.

Then, by Bernoulli’s Inequality,

cn = (1 + x)n ≥ 1 + nx = 1 + x + (n − 1)x ≥ 1 + x = c.

Furthermore, n − 1 > 0 and (n − 1)x > 0 if n > 1.

In that case, the last inequality above is strict and so cn > c > 1,
which implies cn > 1 by transitivity of >.

P. Boily, A. Smith (uOttawa) 2


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

(b) If 0 < c < 1, there exists b > 1 such that c = 1b . Indeed, 1c is such that
c · 1c = 1. As c > 0, then 1c > 0 since the product c · 1c = 1 is positive.

But c < 1, so that 1 = c · 1c < 1c .

In particular, if we let b = 1c , then b > 1 and so we can apply


part (a) of this question to get bn ≥ b for all n ∈ N and bn > 1 if
n > 1.

Let n ∈ N. Then
1 n 1
= b ≥ b =
cn c
so that c ≥ cn and
1 n
n
= b >1
c
so that 1 > cn if n > 1. 

P. Boily, A. Smith (uOttawa) 3


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

3. Let c > 0 be a real number.


(a) If c > 1 and m, n ∈ N, show that cm > cn if and only if m > n.
(b) If 0 < c < 1 and m, n ∈ N, show that cm > cn if and only if m < n.

P. Boily, A. Smith (uOttawa) 4


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

Proof.
(a) It is sufficient to show that if m ≥ n, then cm ≥ cn. (Why is this the
case? Don’t let this slip by without understanding.)

If m = n, the result is clear. So we consider m > n.

In this case, ∃k ≥ 1 such that m = n + k. An easy induction exercise


shows that cn+k = cnck for for all integers n and k (from this point on,
we will assume and apply freely all the usual techniques of algebra).

In particular, using the previous problem,

cm = cn+k = cnck ≥ cn · c > cn · 1 = cn

and so cm > cn.

P. Boily, A. Smith (uOttawa) 5


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

(b) This can be shown from part (a) using the technique from the previous
question. 

P. Boily, A. Smith (uOttawa) 6


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

4. Let S2 = {x ∈ R | x > 0}. Does S2 have lower bounds? Does S2


have upper bounds? Does inf S2 exist? Does sup S2 exist? Prove your
statements.

P. Boily, A. Smith (uOttawa) 7


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

Proof.
Does S2 have lower bounds? Yes.
By definition, any negative real number is a lower bound (so is 0).

Does S2 have upper bounds? No.


Assume that it does. By the completeness of R, α = sup R exists.
In particular, α ≥ n for all n ∈ N, which contradicts the Archimedean
Property of R. Hence S2 has no upper bound.

Does inf S2 exist? Yes.


Consider the set −S2 = {x ∈ R | −x ∈ S2} = {x ∈ R | x < 0}. By
construction, 0 is an upper bound of −S2. Note furthermore that neither
S2 nor −S2 are empty.

By completeness of R, sup(−S2) exists. Right?

P. Boily, A. Smith (uOttawa) 8


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

One definition of completeness is that any non-empty bounded subset of


R has a supremum. But −S2 is only bounded above, not below. How
can we conclude that sup(−S2) exists?

That definition is one particular version of the Completeness Property


of R. An equivalent way of stating it is: The ordered set F is
complete if for any ∅ 6= S ⊂ F , S has a supremum in F whenever
S is bounded above and an infimum in F whenever S is bounded below.

But sup(−S2) = − inf S2. Indeed, let u = sup(−S2). Then u ≥ −x for


all −x ∈ −S2 and if v is another upper bound of −S2 then u ≤ v.

Note that if v is an upper bound of −S2, then v ≥ −x for all −x ∈ −S2,


i.e. −v ≤ x for all x ∈ S2: as a result, −v is a lower bound of S2.

P. Boily, A. Smith (uOttawa) 9


MAT 2125 – Elementary Real Analysis Exercises – Solutions – Q2-Q4

Similarly, if −v is a lower bound of S2, v is automatically an upper bound


of −S2. Then any lower bound of S2 is of the form −v, where v is an
upper bound of −S2.

Then, −u ≤ x for all x ∈ S2 and −v ≤ −u whenever −v is a lower


bound of S2. Hence −u = inf S2 and so u = − inf S2.

As sup(−S2) = − inf S2 exists, so does inf S2.

Does sup S2 exist? No.


See second item. 

P. Boily, A. Smith (uOttawa) 10

You might also like