CALCULUS OF SEVERAL

VARIABLES

BY

Dr. JOYDEEP SENGUPTA

DEPARTMENT OF MATHEMATICS
UNIVERSITY OF GOUR BANGA,
P.O. MOKDUMPUR, MALDA, 732103,
WEST BENGAL, INDIA
E-mail : [email protected]
 Chapter 1

Functions on Rn

Module 1

Meaning of Rn

1
 By the symbol R we denote the collection of all real numbers equipped with
(i) its usual order,
(ii) its field properties with respect to usual addition and multiplication,

(iii) its usual topology inherited from the Euclidean metric, (that is, the dis-
tance between two points x and y in R is given by modulus of their differ-
ence.)

Let n be any natural number. By the symbol Rn we shall mean the following:
If n = 1, then R1 is nothing but R.
If n > 1, then Rn means the n-fold Cartesian product of R. That is,

Rn := {(x1 , x2 , . . . , xn ) : x1 , x2 , . . . , xn ∈ R}.

We usually refer R, R2 , R3 as the real line, the real plane and the real space.
We know that Rn forms a real vector space (i.e., a vector space over the
field R) with respect to component wise addition and scalar multiplication as
described below:
For x = (x1 , x2 , . . . , xn ), y = (y1 , y2 , . . . , yn ) ∈ Rn and any α ∈ R,

(x1 , x2 , . . . , xn ) + (y1 , y2 , . . . , yn ) := (x1 + y1 , x2 + y2 , . . . , xn + yn ),

α(x1 , x2 , . . . , xn ) := (αx1 , αx2 , . . . , αxn ).

The additive identity is θ = (0, 0, . . . , 0), which we refer as ‘the zero vector’ or
‘the origin’. Corresponding to any x = (x1 , x2 , . . . , xn ) ∈ Rn , its negative is
given by
−x = (−x1 , −x2 , . . . , −xn ).

Now Rn forms a real inner product space with respect to the inner product
defined by
hx, yi := x1 y1 + x2 y2 + . . . + xn yn
for all x = (x1 , x2 , . . . , xn ), y = (y1 , y2 , . . . , yn ) ∈ Rn .
This operation is also known as dot product and in that case we denote the
product by putting a dot between the vectors (as x · y denotes the dot product
of the vectors x and y). In case of R, inner product is nothing but the usual
product of two reals.
One can easily check that the inner product on Rn satisfies the following
properties:
(i) hy, xi = hx, yi,
(ii) hcx, yi = chx, yi,

2
(iii) hx + y, zi = hx, zi + hy, zi
for all x, y ∈ Rn and c ∈ R. These properties of the inner product are known as
symmetry, homogeneity and additivity properties.
Two vectors x and y in Rn are said to be orthogonal or perpendicular if and
only if
hx, yi = 0.

The norm on Rn induced by this inner product is given by

1 1
kxk := hx, xi 2 = (x21 + x22 + . . . + x2n ) 2 ,

for all x = (x1 , x2 , . . . xn ) ∈ Rn .

It is a matter of routine verification that

(i) kxk ≥ 0, and kxk = 0 if and only if x = θ;

(ii) kcxk = |c| kxk for any c ∈ R and x ∈ Rn ,
(iii) kx + yk ≤ kxk + kyk for any x, y ∈ Rn .

In case n = 1, i.e. in R1 , kxk reduces to |x|, i.e. the absolute value of x.

Looking at the case for n = 2 we can easily understand that it generalises the
notion of length of a vector in an arbitrary vector space and it’s origin is the
Pythagoras Theorem. Moreover, any Inner Product Space is a Normed Linear
Space, but the converse is not true.
The Cauchy-Schwarz inequality given by

|hx, yi| ≤ kxkkyk, for all x, y ∈ Rn .

The equality holds if and only if any of the two vectors is a scalar multiple of
the other. The proof of this inequality may be found in any standard text book
of Linear Algebra.

A non-empty set X is said to be a metric space if and only if there is

associated a function d : X × X → R satisfying the following properties:
(i) d(x, y) ≥ 0,
(ii) d(x, y) = 0 if and only if x = y,
(iii) d(x, y) = d(y, x),
(iv) d(x, z) ≤ d(x, y) + d(y, z)

3
for all x, y, z ∈ X. This function d is called a distance function or a metric on
X.
One can easily check that Rn forms a metric space with respect to the metric

d(x, y) := kx − yk for all x, y ∈ Rn .

This metric is known as the usual metric or the Euclidean metric on Rn . The
topology on Rn induced by the metric is called ‘the usual topology on Rn ’.
While studying differentiation in R we always considered functions defined
over open intervals (or their supersets). In order to extend the theory in higher
dimensions we require generalisations of open intervals, which are known as
open spheres. Thus our aim now is to introduce the concept of openness in Rn .
We proceed as follows:
Let a be an arbitrary point in Rn and r be any positive real. By ‘the Open
Sphere in Rn with centre at a and radius r’ we mean the collection of all points
in Rn whose distance from the point is (strictly) less than r. The popular
notations for this open sphere is S(a, r) or B(a, r). Here S stands for sphere
and B stands for Ball, another accepted nomenclature for the same. Thus for
any a = (a1 , a2 , . . . , an ) ∈ Rn , we have

S(a, r) := {x ∈ Rn : kx − ak < r}
n p o
= (x1 , x2 , . . . , xn ) ∈ Rn : (x1 − a1 )2 + . . . + (xn − an )2 < r .

The corresponding closed sphere (or closed ball) is denoted by S[a, r] or B[a, r]
and is defined as follows:

S[a, r] := {x ∈ Rn : kx − ak ≤ r}
n p o
= (x1 , x2 , . . . , xn ) ∈ Rn : (x1 − a1 )2 + . . . + (xn − an )2 ≤ r .

An interesting reader can justify himself that an open sphere reduces to

(i) the open interval (a − r, a + r) in R (i.e. for n = 1),

(ii) the disk of radius r and centre at a in R2 ,

(iii) the solid sphere (in literary sense) of radius r and centre at a in R3 .
Let A be an non-empty set in Rn . A point a ∈ A is said to be an ‘interior
point’ of A (or, equivalently we can find some real r), if and only if we can find
an open sphere in Rn with centre at a that lies entirely within the set A, i.e.

a ∈ S(a, r) ⊂ A.

If this is the essential qualification ‘for a point to lie inside a set’, one can
easily guess the condition a point must possess ‘for being outside a set’. We
shall call a point b, an exterior point of a set A, if and only if we can find an

4
open sphere with centre at b (similarly, a suitable real r) such that it shares no
common point with the set A, that is

A ∩ S(b, r) = ∅.

The collection of all interior points of a set is called the interior of the set A
and is denoted by Int(A). The exterior of a set A is denoted by Ext(A) and is
similarly defined.
A set A in Rn is said to be an open set if and only if all its points are interior
points, i.e. Int(A) = A. As the empty set has no point which is not an interior
point, it is also considered to be open.
We have the following:
(i) any open sphere is an open set,
(ii) Rn is open,
(iii) the empty set is open,
(iv) arbitrary union of open sets is open,
(v) intersection of any two open sets is open.
Clearly the last property can be extended for finitely many open sets.
By a neighbourhood of a point a ∈ Rn , we mean any superset of any open
set containing the point a. Clearly, any open sphere or any open set containing
the point a is a neighbourhood of itself.
Now we are going to define a limit point of a set. Let A ⊂ Rn be any set. A
point a is said to be a limit point of the set A, if and only if every neighbourhood
of the point a contains points of the set A different from a, i.e. points of the set
A \ {a}.
A limit point of a set may or may not belong to that set.
The collection of all limit points of a set A is called the derived set of A and
is usually denoted by D(A).
A set is said to be closed if and only if it contains all its limit points.

The following properties regarding closed sets are to be noticed:

(i) any closed sphere is a closed set,
(ii) Rn is closed,
(iii) the empty set is closed,
(iv) arbitrary intersection of closed sets is closed,
(v) union of any two (and hence finitely many) closed sets is closed,

5
(vi) the complement of an open set is closed and the complement of a closed
set is open.

Finally, one more topological property needs our attention, i.e. connected-
ness. The name suggests that a set can be called connected if it consists of a
single piece. The formal definition follows:
A set A is said to be connected if and only if it can never be expressed as
disjoint union of two non-empty open sets.
Equivalently we can replace the word ‘open’ in the definition by ‘closed’.
A finite set in Rn is connected if and only if it is a singleton set. Union or
intersection of two connected sets may not be connected. Union of two connected
sets is connected provided they have non-empty intersection.