REPUBLIQUE ALGERIENNE DEMOCRATIQUE ET POPULAIRE

MINISTERE DE L’ENSEIGNEMENT SUPERIEURE ET DE LA RECHERCHE

SCIENTIFIQUE

Ecole Nationale Polytechnique Constantine

Preparatory Class

Analysis
Course designed for first-year students.
Written by:
Bahi Oussama
Contents

1 Real sequences 5
1.1 Real Sequences: General Concepts . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Convergence of real sequences . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3 Recursive Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
List of Figures

1.1 Bounded sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.2 Convergence of real sequences . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3 Example about squeeze theorem . . . . . . . . . . . . . . . . . . . . . . . . 13

3
List of Tables

4
Chapter 1

Real sequences

For aspiring engineers, understanding the mathematical foundations that underpin com-
plex systems is essential. Among these foundations are real sequences. These aren’t just
abstract mathematical concepts; they find applications in signal processing, systems anal-
ysis, algorithm design, and more. This chapter delves into the world of real sequences,
starting with their general concepts to set the stage. We’ll then explore the pivotal idea of
convergence. Finally, we’ll tackle recursive sequences, often seen in iterative methods and
feedback systems. By the end of this chapter, you’ll have a firm grasp of real sequences
and their relevance in engineering scenarios.

1.1 Real Sequences: General Concepts

Definition 1.1.1.
Let E be an arbitrary set. We call a "sequence" of elements from E any function U defined
on a subset A ⊆ N with values in E, which is to say:

U :A⊆N → E
n 7→ U (n)

Remark 1.1.1.

• It is customary to write Un instead of U (n) (the image of n by U ).

• Un is called the "general term of the sequence (Un )n∈A ."

• (Un ) is called the sequence.

 • If E = R, then the sequence (Un ) is said to be real or numeric.

• Sequences can be defined in two different ways:

1. Directly through a formula, typically a function f , where ∀n ∈ N, we have

Un = f (n). This is known as an explicit formulation of the sequence.
2. By expressing Un+1 in terms of the previous term Un and defining an initial
value, such as:
Un+1 = f (Un )
(

U0 = θ (θ is a known value).

This is referred to as a recursive formulation.

Example 1.1.1.

1. Consider the sequence (Un ) for ≥ 0 with the general term Un = 2

n+1
. Therefore, the
first terms are U0 = 2, U1 = 1, U2 = 23 , and so on.
 n
2. Consider the sequence (Vn ) for n ≥ 1 such that Vn = 1 + 1
n
. Then, V1 = 2,
V2 = 94 , and so on.

3. The sequence (Wn ) is defined by recurrence as follows:


W
n+1= Wn2 + 3,
W0 = 1.

We calculate the terms of the sequence recursively. Therefore, W1 = W02 + 3 = 4,

W2 = W12 + 3 = 19, and so on.

Definition 1.1.2 (Bounded sequences).

Let (Un )n∈N be a numeric sequence.

• We say that (Un )n∈N is upper bounded if:

∃M ∈ R : ∀n ∈ N, Un ≤ M

6
 • We say that (Un )n∈N is lower bounded if:

∃m ∈ R : ∀n ∈ N, Un ≥ m

• A sequence is said to be bounded if it is both upper bounded and lower bounded,

which means:
∃m, M ∈ R : ∀n ∈ N, m ≤ Un ≤ M
or equivalently,
∃c > 0, ∀n ∈ N : |Un | ≤ c

Additionally,

∀n ∈ N : Un ≤ M,
(
M = sup Un ⇔
∀ϵ > 0, ∃n0 ∈ N : M − ϵ < Un0 ≤ M.

∀n ∈ N : m ≤ Un ,
(
M = inf Un ⇔
∀ϵ > 0, ∃n0 ∈ N : m ≤ Un0 < m + ϵ.

Figure 1.1: Bounded sequence

7
Definition 1.1.3 (Variation of a Sequence).

Let (Un ) be a sequence. Then,

1. (Un ) is constant if Un = U0 , ∀n ∈ N.

2. (Un ) is stationary if there exists k ∈ N such that Un = Uk , ∀n ≥ k.

3. (Un ) is increasing if ∀n ∈ N, Un+1 ≥ Un .

4. (Un ) is strictly increasing if ∀n ∈ N, Un+1 > Un .

5. (Un ) is decreasing if ∀n ∈ N, Un+1 ≤ Un .

6. (Un ) is strictly decreasing if ∀n ∈ N, Un+1 < Un .

7. (Un ) is monotonic if it is either increasing or decreasing. (Un ) is strictly monotonic

if it is either strictly increasing or strictly decreasing.

Example 1.1.2.

 
1. The sequence Un = 1
n n≥1
is strictly decreasing over N∗ because

1
Un+1 − Un = − < 0, ∀n ∈ N∗ .
n(n + 1)

2. The sequence Vn = (n2 )n≥1 is strictly increasing over N because

Vn+1 − Vn = 2n + 1 > 0, ∀n ∈ N.

(−1)n
 
3. However, the sequence Wn = n
is neither increasing nor decreasing over
n≥1

N∗ (we can say it is alternating).

Definition 1.1.4 (Arithmetic Sequence).

An arithmetic sequence is any sequence (Un ) for which there exists r ∈ R, called the
common difference, such that:

∀n ∈ N : Un+1 = Un + r.

8
Definition 1.1.5 (Geometric Sequence).
A geometric sequence is any sequence (Un ) for which there exists q ∈ R, called the common
ratio, such that ∀n ∈ N, we have
Un+1 = q · Un .

It is possible to express the explicit formula for each of these sequences using the following
proposition.

Proposition 1.1.1 (Explicit Formulation of Arithmetic and Geometric Sequences).

1. The general term of an arithmetic sequence Un with common difference r and initial
term U0 is given by
Un = U0 + nr.

2. The general term of a geometric sequence Un with common ratio q and initial term
U0 is given by
Un = U0 · q n .

3. For an arithmetic sequence (Un ) with a common difference r and initial term U0 ,
we have:
n−1
n
Uk = (U0 + Un ) .
X

k=0 2

4. For a geometric sequence (Un ) with a common ratio q and initial term U0 , we have:

n−1
qn − 1
!
Uk = U0
X
.
k=0 q−1

Definition 1.1.6 (Subsequences).

Let (Un )n∈N be a numeric sequence. Let ϕ : N → N be a strictly increasing sequence of
natural numbers, meaning that ∀n ∈ N : ϕ(n) < ϕ(n + 1). The sequence (Vn )n∈N defined
as
∀n ∈ N, Vn = Uϕ(n)

is called a "subsequence" or "extracted sequence" of (Un )n∈N .

9
Example 1.1.3.
Consider the sequence (Un )n∈N defined as:

U : N∗ → R,
n 7→ Un = n1 .

The sequences (U2n ), (U2n ), (U2n+1 ), and (U3n ) are subsequences of the sequence (Un )n∈N .

1.2 Convergence of real sequences

Definition 1.2.1.
Let (Un )n∈N be a numeric sequence. We say that (Un ) is convergent and has a limit l ∈ R
if it satisfies the following property:

∀ϵ > 0, ∃n0 ∈ N, ∀n ∈ N : n ≥ n0 =⇒ |Un − l| < ϵ

We denote this as:

lim Un = l,
n→+∞

or equivalently,

lim Un = l ⇐⇒ [∀ϵ > 0, ∃n0 = n0 (ϵ) ∈ N, ∀n : n ≥ n0 =⇒ |Un − l| < ϵ] .

n→+∞

A sequence that is not convergent is called divergent (if its limit is infinite or does not
exist).

10
 Figure 1.2: Convergence of real sequences

Example 1.2.1.
Consider the numeric sequence (Un )n∈N defined as follows:

−2n + 1
Un =
n+1

using the definition of convergence show that:

lim Un = −2.
n→+∞

Theorem 1.2.1 (Uniqueness of the Limit).

If (Un ) is a convergent sequence with limit l, then the limit l is unique.

Example 1.2.2.

The sequences Un = (−1)n , Vn = sin n, and Wn = cos n are not convergent because
they do not have unique limits for each of them.

Theorem 1.2.2.
If (Un ) is a convergent sequence, then (Un )n∈N is bounded.

11
Remark 1.2.1.
The converse is not always true, meaning a bounded sequence is not necessarily convergent.
However, if it is not bounded, then it is divergent.
Example 1.2.3.

The sequence Un = (−1)n is bounded, but it is not convergent because it does not have
a unique limit.
Theorem 1.2.3.

If the sequence (Un ) is convergent, then any subsequence of (Un ) converges to the same
limit as (Un ).

Remark 1.2.2.
This result is used to demonstrate that a sequence is divergent. Specifically, if we find
a divergent subsequence or two subsequences that converge to different limits, then the
original sequence is divergent.
Theorem 1.2.4.
Let (Un ) and (Vn ) be two sequences converging to l and l0 respectively. Then, the sequences
(Un + Vn ), (Un · Vn ), and (αUn ), where α ∈ R, converge to l + l0 , ll0 , and αl respectively.
Similarly, the sequence
Un
 
if Vn ̸= 0,
Vn
converges to l
l0
if l0 ̸= 0.

Proposition 1.2.1.

If lim Un = l then lim |Un | = |l|

n→+∞ n→+∞

Proposition 1.2.2 (Extension of Inequalities).

If lim Un = l and , ∀n > n0 : Un > 0, then l > 0.

n→+∞

Corollary 1.2.1.
Let (Un ) and (Vn ) be two convergent numeric sequences such that for n sufficiently large

If Un ≤ Vn ⇒ lim Un ≤ lim Vn .
n→+∞ n→+∞

12
Theorem 1.2.5 (Squeeze Theorem).
If two numeric sequences (Un ) and (Vn ) converge to the same real number l, and for some
index, the sequence (Wn ) satisfies

Un ≤ Wn ≤ Vn

then (Wn ) also converges to l.

Figure 1.3: Example about squeeze theorem

Example 1.2.4.

sin n
lim
Calculate n→∞
n
.
Proposition 1.2.3.
Let (Un )n∈N be a bounded sequence, and let (Vn )n∈N be a sequence that converges to 0.
Then, the sequence (Un Vn )n∈N converges to 0.

Example 1.2.5.
(−1)n
Un = sin n
n
and Vn = n
for n ∈ N.

Previous results do not extend to non-convergent sequences. Among divergent se-

quences, one distinguishes those whose general term tends to infinity.

13
Definition 1.2.2.
We say that a sequence (Un ) tends towards +∞ (resp. −∞) if:

∀A > 0, ∃n0 ∈ N : ∀n ≥ n0 ⇒ Un > A (resp. Un < −A).

We denote this as:

lim Un = +∞ (resp. lim Un = −∞).
n→+∞ n→∞

These previous results extend in certain cases to infinite limits. For example, if:

lim Un = +∞ and lim Vn = +∞, then : lim (Un + Vn ) = +∞.

n→+∞ n→+∞ n→+∞

On the other hand, if:

lim Un = +∞ and lim Vn = −∞ then nothing can be said about lim (Un + Vn )
n→∞ n→+∞ n→∞

as it is an indeterminate form.

Theorem 1.2.6.
We have the following:
√
1. limn→∞ n
a = 1 for a > 0.
√
2. limn→∞ n
np = 1 for all p ∈ N.

0 |q| < 1,



 if
1 q = 1,

if

3. limn→∞ q n =




Does not exist if q = −1,
1 |q| = 1.


if

nk
4. limn→∞ an
= 0 for k ∈ N and a > 1.

an
5. limn→∞ n!
= 0 for a ∈ R.

6. limn→∞ n!
nn
= 0.
 n
7. limn→∞ 1 + 1
n
= e for 2 < e < 3.

14
Example 1.2.6. Compute:
!n
3n2 + 5n + 1
lim .
n→∞ 3n2 + 3n + 1

Theorem 1.2.7.
Let (Un ) be a numerical sequence.

1. If (Un ) is increasing and upper bounded, then (Un ) is convergent.

2. If (Un ) is decreasing and lower bounded, then (Un ) is convergent.

3. If (Un ) is increasing and not upper bounded (respectively, decreasing and not lower
bounded ), then
lim Un = +∞
n→∞

respectively,
lim Un = −∞
n→∞

Example 1.2.7.
 n
Prove that the sequence Un = 1 + 1
n
is convergent.

Definition 1.2.3 (Adjacent Sequences).

We say that two numerical sequences (Un ) and (Vn ) are adjacent if they satisfy the fol-
lowing three conditions:

1. (Un ) is increasing.

2. (Vn ) is decreasing.

3. limn→∞ (Un − Vn ) = 0.

Theorem 1.2.8.
If (Un ) and (Vn ) are two adjacent sequences, then they are convergent and have the same
limit.

15
Definition 1.2.4 (Cauchy Sequences).
Let (Un ) be a numerical sequence. (Un ) is said to be a Cauchy sequence if:

∀ϵ > 0, ∃n0 ∈ N, ∀p, q ∈ N : p > q ≥ n0 =⇒ |Up − Uq | < ϵ.

Negation: A sequence (Un ) is not Cauchy if:

∃ϵ > 0, ∀n0 ∈ N, ∃p, q ∈ N : p > q ≥ n0 and |Up − Uq | ≥ ϵ.

Theorem 1.2.9. If (Un ) is convergent, then (Un ) is a Cauchy sequence.

Example 1.2.8. Let the sequence Un be defined as Un = 1 + 1

2
+ 1
3
+ . . . + n1 ; study its
nature.

Theorem 1.2.10. If (Un ) is a Cauchy sequence of real numbers, then (Un ) is convergent.

1.3 Recursive Sequences

Definition 1.3.1.
Let f : D ⊆ R → R be a function such that f (D) ⊆ D:

• A "recursive sequence" is any sequence (Un )n∈N defined by the initial term U0 ∈ D
and the relation Un+1 = f (Un ) for all n ∈ N.

• Monotonicity: The study of the monotonicity of the recursive sequence (Un )n∈N
defined by:

 U0 ∈ D,
 Un+1 = f (Un ).

is equivalent to that of the function f . Using the recurrence, with

Un+1 − Un = f (Un ) − f (Un−1 ),

we find that:

1. If f is increasing, then (Un ) is monotonic. More precisely,

– (Un ) is increasing if, in addition, U0 ≤ U1 .

– (Un ) is decreasing if, in addition, U0 ≥ U1 .

16
 • Convergence: Suppose f is continuous on D. If the recursive sequence (Un ) con-
verges to l in D, then we have: f (l) = l. The search for the limit then boils down
to the study of the equation f (l) = l.

Example 1.3.1. Study the nature of the recursive sequence defined by

=√
0
(
U0
Un+1 = 6 + Un , n ≥ 0.

17