MA 105

Calculus of one and

several variables
Prof. Inder K. Rana

Room 112 B
Department of Mathematics
IIT-Bombay, Mumbai-400076 (India)
Email: [email protected]
Lecture 1

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Basics of Calculus

Rigorous foundation of calculus, needs understanding of the

following concepts:
The concept of a real number.
The concept of a function.
The concept of limit.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Real numbers

Real numbers are the elements of a set, denoted by IR, with the
following properties:
Algebraic properties of real numbers:
There are two binary operations defined on IR, one called addition,
denoted by
(x, y ) 7−→ x + y ,
the other called multiplication, denoted by
(x, y ) → xy ,
with the usual algebraic properties:

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Real numbers

for all x, y , z ∈ IR,

(i) x + y = y + x; xy = yx (commutative law).
(ii) x + (y + z) = (x + y ) + z; x(yz) = (xy )z (associative law).
(iii) x(y + z) = xy + xz; (y + z)x = yx + zx (distributive law).
(iv) There exist two distinct elements in IR, denoted by 0 and 1, with
properties:
0 + x = x for all x ∈ IR;
1x = x for all x ∈ IR.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Real numbers

The element 0, read as zero, is called the additive identity,

and the element 1, read as one, is called the multiplicative identity.
(v) For every x ∈ IR there exists unique element −x ∈ IR such that
x + (−x) = 0;
for x 6= 0 in IR, there exists unique element x −1 ∈ IR, such that
xx −1 = 1.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Real numbers

Order properties of real numbers:

There exists an order, denoted by <, between the elements of IR
with the following properties:
(i) For x, y ∈ IR, one and only one of the following relations hold:
x < y , x = y , y < x.
(ii) x, y > 0 implies xy > 0 and x + y > 0.
(iii) x > y and y > z implies x > z.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Real numbers

(3) Archimedean property:Given x ∈ IR, x > 0, for every y ∈ IR there

exists n ∈ IN such that n x > y .
There is one more (important ) property that real numbers have
which we shall describe later:
(4) Completeness property.
There are some special subsets of IR which are important. These
are the familiar number systems.
• IN = {1, 2, . . . , }, called the set of natural numbers.
• ZZ := {. . . , −2, −1, 0, 1, 2, . . .}, the set of integers.
I := {nm−1 | n ∈ ZZ , m ∈ IN}, the set of rational numbers.
• Q

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Geometric representation of real numbers

| |

| |
1 2

| | | | |
1 2

| | | | |
−2 −1
Prof. Inder K. Rana 0
Department 1
of Mathematics, IIT - 2
Bombay
Functions

We shall assume that you are familiar with the concept of

intervals in IR, and the concept of functions and their properties.
Read sections 1.1.6 - 1.2.11 from "Calculus @ iitb"
The absolute value function | · | : IR −→ [0, +∞) is defined by

x if x ≥ 0,
|x| :=
−x if x < 0,

|x| is called the absolute value of x ∈ IR.

Geometrically, |x| is the distance of x from 0.

and |x − y | is the distance between x and y .

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Properties of the absolute value function

Theorem
For every x, y , z ∈ IR :
(i) |x| ≥ 0 and |x| = 0 if and only if x = 0.
(ii) |x| = | − x|.
(iii) |xy | = |x| |y |.
(iv) − | x | ≤ x ≤ | x |. Proof

(v) |x| ≤ y if and only if −y ≤ x ≤ y . Proof

(vi) |x + y | ≤ |x| + |y |, (triangle inequality). Proof

(viii) | |x| − |y | | ≤ |x − y |. Proof

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Difference between IR and Q
I

What do you think are the similarities and differences between

real numbers and rational numbers?
1 Both are "ordered fields".
2 I is a proper subset of IR.
Q
3 I is a countably infinite whereas IR is uncountable.
Q
4 IR is a "complete" whereas Q
I is not.
We shall describe the last property in brief.
For details, if interested, have a look at the book:
"From Numbers to Analysis".

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Need for sequences
Any iterative process gives rise to a sequence of observations.
For example

Let An denotes the area of the n side polygon inscribed in the

circle, this gives us the sequence of numbers A3 , A4 , . . . , An . . .
Prof. Inder K. Rana Department of Mathematics, IIT - Bombay
Sequences

A sequence of elements of a set S is an ordered collection:

a1 , a2 , . . . , an . . . , of elements of S.
The number a1 is called its first term, a2 - its second term, and in
general an as its nth-term.
We usually denote a sequence a1 , a2 , . . . , an . . . , as {an }n≥1 .
Note that {an }n≥1 is not same as the set {an |n ≥ 1}.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Sequences

Example
(i) {an }n≥1 , an = population of India in the year n.
(ii) For {(−1)n }n≥1 , a1 = −1, a2 = 1, a3 = −1, . . . .
(iii) For {cos n π}n≥1 , a1 = −1, a2 = 1, a3 = −1, . . . .
 
1
(iv) For ,a1 = 1, a2 = 12 , a3 = 13 , . . . .
n n≥1

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

Given a sequence {an }n≥1 , one is interested to know:

what happens to an as n becomes large?
Do an0 s come close to some value?
For example, the numbers An , the areas of regular n-gons
inscribed in the unit circle come closer to ’area’ of the unit circle.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

convergence of sequences

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

convergence of sequences
A sequence {an }n≥1 is said to converge if there exists ` ∈ R such
that

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences
Geometrically, a sequence is convergent ` if every small
neighborhood of ` includes a ’tail’ of the sequence.

A sequence which is not convergent is called a divergent

sequence.
As a consequence of the Archimedean property of real numbers
we have the following:
Prof. Inder K. Rana Department of Mathematics, IIT - Bombay
Convergent Sequences

• Examples:
1
(i) The sequence n n≥1 is convergent to ` = 0.
Let  > 0 be given. Then
|an − `| =  n1 − 0 = n1 .
1
As n <  iff n > 1 ,
we choose(using the Archimedean property) n0 , such that
n0  > 1, i.e., n0 > 1/. Then,
1 1
for all n ≥ n0 , n ≤ n0 < .
Hence,
|an − `| = | n1 − 0| = 1
n <  for all n ≥ n0 ,
1
i.e., lim = 0.
n→∞ n

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

• Example:
The sequence {n}n≥1 is not convergent.
Suppose, lim n = `. Then,
n→∞
there exists n0 ∈ N such that
|an − `| = |n − `| < 1, for all n ≥ n0 ,
i.e.,
` − 1 < n < ` + 1 for all n ≥ n0 ,
which contradicts the Archimedean property.
Hence, {n}n≥1 is divergent.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

• Example:
Consider the sequence {(−1)n }n≥1 .
Every odd term of the sequence is −1 and every even term of the
sequence is +1.
We expect {(−1)n }n≥1 to be divergent.
Let us prove this.
First suppose that an → l, l = −1.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

• Note:
Though both sequence {n}n≥1 and {(−1)n }n≥1 are divergent, they
are divergent for different reasons.
{n}n≥1 is divergent as it keeps growing.
{(−1)n }n≥1 is divergent as it oscillates between −1 and 1.
• Example:  
n
Consider the sequence .
n+1 n≥1

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

Since
n (n + 1) − 1 1
= =1− ,
n+1 n+1 
 n+1
n
we guess that lim = 1.
n→∞ n + 1
As  
 n
= 1 ,


n + 1 − 1 n+1
1
for any given  > 0, if we select n0 such that ≤ , then
n0

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Convergent Sequences

for all n ≥ n0 − 1, 
 n  1 1
 n + 1 − 1 = n + 1 < n < .
 
0
Hence,  
n
lim = 1.
n→∞ n + 1
Question: Can a sequence have more than one limit?
Intuitive answer: NO! Proof

Since a sequence which grows as large as we want, cannot

converge, we have next concept:

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Bounded Sequences
• Definition :
Let {an }n≥1 be a sequence.
(i) {an }n≥1 is bounded above if for some α,
an ≤ α for all n.

(ii) {an }n≥1 is bounded below if an ≥ β for all n for some β.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Bounded Sequences

(iii) {an }n≥1 is bounded if it is bounded both above and below.

• Examples:
n n
o
(i) (−1)
n 2 is bounded, since for all n,
n≥1 n 
 (−1)  1
 n2  = n2 < 1.
 
 n
2
(ii) The sequence is not bounded, since
n2 n≥1
2n > n2 > n for all n ≥ 5.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Bounded Sequences

Theorem
If {an }n≥1 is convergent, then it is bounded.

• Examples:
The sequence {(−1)n }n≥1 in bounded but not not convergent.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Limit theorems : Algebra of limits

Theorem
Let {xn }n≥1 and {yn }n≥1 be sequences such that
lim xn = x and lim yn = y .
n→∞ n→∞
Then the following holds:
(i) The sequence {xn + yn }n≥1 is convergent and
lim (xn ± yn ) = x ± y .
n→∞

(ii) The sequence {xn yn }n≥1 is convergent and

lim (xn yn ) = xy .
n→∞

(iii) If y 6= 0, then there exists some

n no1 ∈ N, such that xn /yn is defined for for
xn
all n ≥ n1 , and the sequence yn , is convergent with
n≥n1
xn x
lim = .
n→∞ yn y

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Limit theorems

Proofs of (i) and (ii) are reading exercises

(see theorem 3.2.1 of calculus@iitb) and (iii) is assumed without
proof.

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Recap

We discussed:
• The need to consider sequences.
• The notion of a sequence.
• The notion of convergent sequence
• The notion of bounded sequence
• The Archimedean property of real numbers
• Limit theorems

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay

Next

 
Next Topic
 

Prof. Inder K. Rana Department of Mathematics, IIT - Bombay