Calculus

18B11MA111

Course Instructor: Dr R K Bajaj

Differentiation and integration of functions of

two and more variables.
 Functions of several variables
Many functions depend on more than one variable.
Some familiar examples are:
Distance = velocity × time or d=vt
Volume V of a circular cylinder of radius r and height h
V = π r2 h
The diagonal of a rectangle of length l and breadth b
d  l 2  b2

z  x2  y 2
Gives the height of a point on paraboloid above xy-plane.
Definitions:
Let D be a set of n-tuples (x1, x2, x3’ ……, xn ) of real numbers. A real-
valued function f on D is a rule that assigns a real number
w  f ( x1 , x2 , x3 ,......, xn )

to each n-tuple in D.
The set D is called domain.
The set of values w given by the rule f is the range of the function.
The symbol w is the dependent variable defined by f, where f is a
function of n independent variables x1, x2 , x3 , ……, xn .
If f is a function of two independent variables, they are usually taken
as x and y, the domain of f is a region in xy-plane. Dependent
variables is taken as z.
If f is a function of three independent variables, they are usually
taken as x, y and z, the domain of f is a region in space.
In application we use letters which remind us what the variables
represent.
To find the value of dependent variable corresponding to given set of
independent variables we substitute the values of independent
variables in the formula defining the function and calculate
corresponding value of dependent variable.
Example: Find the distance of the point (3, 0, 4) from origin.
Solution:

The distance of a point ( x, y, z) form origin is given by

d  f ( x, y , z )  x 2  y 2  z 2
Therefore the distance of the point (3, 0, 4) from origin.
w  f (3,0, 4)  32  02  42  25  5
Domain & Range:
Domain is the largest set of independent variables which give real
finite and unique value of dependent variable.
Range is the interval which contains all possible values of dependent
variable.
Example:
Function Domain Range
w  x  y2 x  y2 [0, )
1
w xy  0 ( , 0)  (0, )
xy

w  x2  y 2  z 2 Entire space [0, )

 Definitions:
Disk of radius r is the set of points (x, y) which satisfy
x2  y 2  r 2
A point (x0, y0) in a region (or set) R in xy- . ( x0 , y0 )
plane is an Interior point of R if it is the
R
centre of a disk which lies entirely in R.

A point (x0, y0) Boundary point of R if . ( x0 , y0 )

every disk centered at this point contains
points that lie outside R as well as the points R
that lie in R.
All interior points of region R constitute a set called Interior of R.
All their boundary points of region R form its Boundary.
A region is said to be open if all points of region are interior points.
A region is said to be closed if all boundary points of region are
contained in it.
A region is bounded if it lies inside a disk of fixed radius.
Examples:
Bounded regions (sets) in 2D:
Line segments, triangles, interior of a rectangle etc.
Un-bounded regions (sets) in 2D:
Lines, Coordinate axis, Graphs of tan x, Half planes
y
Example:
Consider the function f ( x, y )  y  x Interior
2
y  x2
Its domain is y  x 2 Boundary
y  x2
It is closed and unbounded.
x
O
 Limit
If the values of f(x, y) are arbitrarily close to a fixed real number L
for all points (x, y) sufficiently close to a point (x0, y0), we say f
approaches the limit L as (x, y) approaches (x0, y0).
And we write
lim f ( x, y)  L
( x , y )( x0 , y0 )

δ – ε Definition:
If for every number ε > 0 however small, there exists a corresponding
number δ > 0 such that for all (x, y) in the domain of f

0  ( x  x0 )2  ( y  y0 )2    f ( x, y)  L  
we say f approaches the limit L as (x, y) approaches (x0, y0), and
write lim f ( x, y)  L
( x , y )( x0 , y0 )
The δ – ε requirement in the definition is equivalent to
0  | x  x0 |  , 0  | y  y0 |   f ( x, y)  L  
Thus while calculating limits we may consider the distance in the
plane or distance along coordinate axis.
The point (x, y) must be inside the domain the point (x0, y0) may be
inside the domain or on the boundary.
As in case of functions of single variables we have:
lim x  x0 lim y  y0 lim k k
( x , y )( x0 , y0 ) ( x , y )( x0 , y0 ) ( x , y )( x0 , y0 )

If lim f ( x, y)  L and lim g ( x, y)  M

( x , y )( x0 , y0 ) ( x , y )( x0 , y0 )

then following Properties hold:

1. Sum/Difference Rule : lim [ f ( x, y)  g ( x, y)]  L  M
( x , y )( x0 , y0 )

2. Product Rule : lim [ f ( x, y)  g ( x, y)]  L M

( x , y )( x0 , y0 )

3. Constant Multiple Rule : lim kf ( x, y)  kL

( x , y )( x0 , y0 )

f ( x, y ) L
4. Quotient Rule : lim  , M 0
( x , y ) ( x0 , y0 ) g ( x, y ) M
5. Power Rule If m and n are integers, then
[ f ( x, y )]  L , provided L n is a real number.
m m m
lim n n

( x , y ) ( x0 , y0 )

x sin y
Example: Find lim
( x , y ) (1,0) x 2  1

x sin y
lim 0
( x , y ) (1,0) x 2  1
 x2  y 2
Example: Find ( x , ylim
)(1,1) x  y
x y

Solution: As (x, y) approaches (1, 1) both denominator and

numerator approach 0. Since x ≠ y, the factor x – y can be
cancelled.
x2  y 2
Therefore we have lim  lim ( x  y )  2
( x , y )(1,1) x  y ( x , y ) (1,1)
x y x y

2 xy  yz
Example: Find lim
( x , y , z ) (1, 1, 1) x 2  z 2

2(1)(1)  (1)(1) 1
  
12  (1)2 2
In case of functions of single variables existence of limit is ensured
if right hand limit is equal to left hand limit.
In case of functions of two or more variables, a point can be
approached in more than two ways.
For example in case of functions of two variables some of the
approaches are: Y

.(x, y)
.
(x0, y0)
X
O
Two-Path test for the Non-existence of a limit :
If a function f(x, y) has two different limits along two different paths
as (x, y) approaches (x0, y0), then limit of the function does not
exist.
 x
Example: Show that the function f ( x, y )  has no
limit as (x, y)  (0, 0). x2  y 2

Example: Along the line y = mx

x
f ( x, y ) y  mx 
x2  y 2 y  mx

x 1
 
x  ( mx )
2 2
1  m2

The limit varies with path of approach.

1
If (x, y)  (0, 0) along y = x the limit is
2
1
If (x, y)  (0, 0) along y = 2 x the limit is
5
 Continuity:
A function f(x,y) is continuous at point (x0, y0), if
1. f is defined at (x0, y0),
2. lim f ( x, y) exists
( x , y )( x0 , y0 )

3. lim f ( x, y)  f ( x0 , y0 )
( x , y )( x0 , y0 )

A function f(x,y) is continuous if it is continuous at every point of

its domain
Example: Show that the following function is continuous at every
point of xy-plane except at origin
 2 x2 y
 4 , ( x, y)  (0, 0)
f ( x, y )   x  y 2

0, ( x, y)  (0, 0)

Solution: Along the curve y = k x2 , x ≠ 0, the function f(x, y)
changes to
2 x2 y 2 x 2
( kx 2
) 2kx 4
2k
f ( x, y ) y  kx2  4  4  4 
x  y y  kx2 x  (kx )
2 2 2
x k x 2 4
1  k 2

Therefore
lim f ( x, y )  lim  f ( x, y) 2   2 k
( x , y ) ( x0 , y0 ) ( x , y ) ( x0 , y0 )  y  kx 
1 k 2
along y  kx 2

This limit varies with path of approach.

If (x, y) approaches (0, 0) along the parabola y = x2 , (k = 1), the
limit is 1.
If (x, y) approaches (0, 0) along the parabola y = - x2 , (k = - 1), the
limit is -1.
By two path test, f has no limit as (x, y) approaches (0, 0).
As the limit does not exist, therefore the function is not continuous
at origin.
 Partial Derivatives
Let variable z be dependent on independent variables x, y and be
given by z = f(x, y).
The rate of change of f (or z) with respect to x (keeping y constant)
is called the Partial Derivative of f with respect to x and is
denoted by f x ( x, y).
The rate of change of f (or z) with respect to y (keeping x constant)
is called the Partial Derivative of f with respect to y and is
denoted by f y ( x, y ).
f ( x  x, y )  f ( x, y )
We define f x ( x, y )  lim
x 0 x
f ( x, y  y)  f ( x, y)
f y ( x, y)  lim
y 0 y
Example: Let f ( x, y)  xy 2
Then f ( x  x, y )  f ( x, y )
f x ( x, y )  lim
x 0 x
( x  x) y 2  xy 2 x 2
 lim  lim y  y 2

x 0 x x  0 x

f ( x, y  y)  f ( x, y) x( y  y)2  xy 2
f y ( x, y)  lim  lim
y 0 y y 0 y
x[2 yy  (y)2 ]
 lim  lim[2 xy  y]  2xy
y 0 y y 0

Observe that obtaining derivatives is similar to the case of single

variable functions.
While obtaining partial derivative with respect to x, y is taken
constant, and while obtaining partial derivative with respect to y, x
is taken constant
  xy 2
Example: If f ( x, y)  3x  e 2
find partial derivatives with
respect to x and y.
Solution:
2  xy 2
f x ( x, y )  6x  y e f y ( x, y )  2 xye  xy 2

Example: Find the partial derivatives with respect to x and y of

cos( xy )
f ( x, y ) 
y
Solution: f x ( x, y )   sin( xy )
 x sin( xy) y  cos( xy).1 xy sin( xy )  cos( xy )
f y ( x, y )   
y2 y2

Example: Find the partial derivatives with respect to x, y and z of

f ( x, y, z )  zx 2  3xy 4
Solution:
f x ( x, y, z )  2 zx  3 y 4 f y ( x, y, z)  12 xy3 f z ( x, y , z )  x 2
Notation:
If z  f ( x, y )
Then partial derivative f x ( x, y ) can be written as

f ( x, y ) z
or f x or
x x

The partial derivative f x ( x, y ) evaluated at (x0, y0) can be expressed

as
f f ( x0 , y0 )
f x ( x0 , y0 ) or or
x ( x , y )
0 0
x

Similar notation is used for partial derivative f y ( x, y ).

 Geometrical meaning of partial derivatives
Suppose the graph of z = f(x,y) is the
surface shown.
Consider the partial derivative of f
with respect to x at a point (x0, y0).
Holding y constant and varying x,
we trace out a curve on the surface
that is the intersection of the surface
with the vertical plane y = y0.
The partial derivative fx(x0,y0)
y = y0
measures the change in z per unit
increase in x along this curve.
That is, fx(x0, y0) is just the slope of
the curve at (x0, y0).
Similar meaning holds for fy(x0, y0) .
 Higher-Order Partial derivatives
For a function f ( x, y ) the partial derivatives
f and f ( x, y )  f
f x ( x, y ) 
y
y
x
are themselves are functions of x and y, so we can obtain their
derivatives.
  f   f
2
  f   f 2

   x 2  f xx    yx  f xy
x  x  y  x 

  f   2 f   f   2 f
   2  f yy    f yx
y  y  y x  y  xy
fxy and fyx are called mixed second-order partial derivatives.
If f, fx, fy, fxy, and fyx are continuous on an open region, then fxy
= fyx at each point in the region.
So the order in which the differentiation is done does not matter.
Further Higher-order partial derivatives (for example fxxy) can also
be calculated.
If the subscript notation is used, the order of differentiation is from
left to right.
Thus we have
    f  
f xxy    
y  x  x  
Example: Find the second partial derivatives of
f ( x, y)  x 2  4 xy 3

Solution:
f x ( x, y )  2 x  4 y 3 f y ( x, y)  12 xy 2

f xx ( x, y)  2 f xy ( x, y)  12 y 2 f yy ( x, y )  24 xy
 x
Example: Find the value of at the point (1, -1, -3) if the
z
equation
xz  y ln x  x 2  4  0
defines x as function of two independent variables y and z and the
partial derivative exists.
Solution: As the equation defines x as function of two
independent variables y and z, therefore differentiating the equation
with respect to z, we get
x 1 x x
zx y  2x  0
z x z z
y x
( z   2 x)  x  0
x z
x x 1

1
 
z y 1 6
z   2x 3   2 1
x 1
 Example: Find the second order partial derivatives of f ( x, y)  x
y

Solution:
 y y 1  y  x y ln x
f x ( x, y )  x  yx f y ( x, y)  x
x y

f xx ( x, y )  yx y 1  y( y  1) x y 2
x

f xy ( x, y )  yx y 1  x y 1  yx y 1 ln x  x y 1 (1  y ln x)
y
 y
f yy ( x, y )  x ln x  x (ln x)
y 2

y
 y y 1 y 1
f yx ( x, y )  x ln x  yx ln x  x  x y 1 (1  y ln x)
x x
 Which variable is taken as constant?
The cartesian coordinates (x, y) and polar coordinates (r, θ) are
related through the relations
x  r cos y  r sin  r 2
 x 2
 y 2
x tan   y
r
If we want to find an expression for
x 2 2 2
then we can use either x  r cos or r  x  y
r
Taking  constant, x  r cos gives  sec
x
Taking y constant, r 2  x2  y 2 gives r x

x r
Observe that two values obtained are not equal.
To avoid confusion we adopt the notation
 r   r  x
   sec   
 x   x  y r
 r  r 
If no indication is given then represents  
x  x  y
Example: If x = r cos θ and y = r sin θ, find
r x  x
, , ,
x r x 
Also find how they are related.
Solution:
r
To find y is taken constant, and we use the expression
x
r r x
r 2  x2  y 2  2r  2 x or 
x x r
x
To find θ is taken constant, and we use the expression
r x
x  r cos   cos 
r
r x x
  cos  
x r r
 1
dw
Example: If w  sin ( x  y), x  3t and y  4t , find dt
3
using
partial derivatives and directly. Show that both procedures give
identical result in terms of t.
Solution: We have
dw w dx w dy
 
dt x dt y dt
dw 1 1
 3 12t 2
dt 1  ( x  y)2 1  ( x  y)2
3  12t 2 3(1  4t 2 )
 
1  (3t  4t ) 3 2
1  9t 2  24t 4  16t 6
3(1  4t 2 ) 3(1  4t 2 )
 
(1  t )(1  8t  16t )
2 2 4
(1  t 2 )(1  4t 2 )2
3

1 t2
Case 2: Differentiation of implicit functions
If w = f(x, y) = c, where c is constant, we have an implicit relation
between x and y.
It defines y a differentiable function of x and the relation
dw w dx w dy
 
dt x dt y dt
takes the form dw w w dy
  0
dx x y dx
w
  x
dy
Hence we have
dx w
y
If w = f(x, y) and x and y are related by an implicit relation g(x, y) =
c, where c is constant, then g
dw w w dy w w x
   
dx x y dx x y g
y
Example:
Find the total differential coefficient of sin( x 2  y 2 ) with respect to x,
when x and y are related by the relation x2  xy  y 2  1.

Solution:
d 2 d
sin( x  y )  cos( x  y ) ( x 2  y 2 )
2 2 2

dx dx
dy
 (2 x  2 y ) cos( x 2  y 2 )
dx
Differentiating the relation x 2  xy  y 2  1 with respect to x, we get
dy dy dy 2x  y
2x  y  x  2 y 0 or 
dx dx dx x  2y
d 2x  y
Therefore sin( x 2
 y 2
)  (2 x  2 y )cos( x 2
 y 2
)
dx x  2y
 x2  y 2 
 2  cos( x 2
 y 2
)
 x  2y 
Example:
At a particular instant base of triangle is 4 cm and height is 3 cm and
they are increasing at the rate of 1.5 cm and 0.5 cm respectively. Find
the rate at which the area is increasing at that instant.
Solution: The area A of triangle with base b and height h is
1
A  bh
2
The rate of change in area
dA 1 db 1 dh
 h b
dt 2 dt 2 dt
At the given instant of time
db dh
b  4cm, h  3cm,  1.5cm / s,  0.5cm / s
dt dt
Therefore dA 1 1
 1.5  3   4  0.5  2.25  1.0  3.25sq cm / s
dt 2 2
Jacobian
If u = (x, y) and v = (x, y) are partially differentiable functions
of x and y, then the determinant
u u
x y u v v u
 
v v x y x y
x y
is called Jacobian of u and v with respect to x and y.
It is denoted by
 (u , v)  u, v 
or J   or J  x, y 
 ( x, y )  x , y 
or simply J, if there is no confusion about the variables involved.
Some Properties Jacobian
 (u, v)  ( x, y )
1. If J and J   then JJ   1.
 ( x, y )  (u, v)
2. If u, v are functions of r, s and r, s are function of x, y then
(u, v) (u, v) (r , s)
 
( x, y) (r , s) ( x, y)
Proof:
1. Let u   ( x, y ) and v   ( x, y ).
Then
u  x  y u  x  y
1  0 
u x u y u v x v y v

v  x  y v  x  y
0  1 
u x u y u v x v y v
Therefore
 (u, y ) ( x, y ) ( , ) ( x, y)
JJ     
( x, y ) (u, v) ( x, y) (u, v)
  x x
x y u v
 
  y y
x y u v

 x  y  x  y
 
x u y u x v y v

 x  y  x  y
 
x u y u x v y v

1 0
 1
0 1
 (u, v) (u, v) (r , s)
2.  
( x, y ) (r , s) ( x, y )
u u r r
(u, v) (r , s) r s x y
  
(r , s) ( x, y) v v s s
r s x y
u r u s u r u s
 
r x s x r y s y

v r v s v r v s
 
r x s x r y s y

u u
x y  (u , v)
 
v v  ( x, y )
x y
 Example: If (x, y) are Cartesian coordinates and (r, θ) are polar
Coordinates of a point in plane, find
 ( x, y )
 ( r , )
Solution: We have
x  r cos y  r sin
x x
 ( x, y ) r  cos r sin 
 
 ( r , ) y y sin  r cos
r 
 r cos 2   r sin 2  r
 ( r , )
?
 ( x, y )
Example: If (x, y, z) are Cartesian coordinates, (ρ, θ, z) are
cylindrical Coordinates and (r, ø, θ) are spherical coordinates of
a point in space, find
( x, y, z )  ( x, y , z )
 (  , , z )  ( r ,  , )
Solution: The coordinate systems are defined as shown in figure.
Z
Therefore we have

x   cos , y   sin  , zz

P

x  r cos sin  , 
r
z

y  r sin  sin  , O Y

z  r cos   ρ x

A y B
X
 x x x
r  
 ( x, y , z ) y y y

 ( r , ,  ) r  
z z z
r  

cos sin  r cos cos  r sin  sin 

 sin  sin  r sin  cos  r cos sin   r 2 sin 
cos  r sin  0

 ( x, y , z )

 (  , , z )
Series Representation of a function of single variable
We know that sin x, cos x, e x
can be represented by infinite power series
x3 x5 x 7
sin x  x     .............
3! 5! 7!
x 2 x 4 x6
cos x  1     .............
2! 4! 6!
2 3 4
x x x
e x  1  x     .............
2! 3! 4!
The functions are continuous and have derivatives of all orders.
These series hold for all values of x.
Near x = 0 the values of functions up to the desired accuracy can be
calculated by taking suitable number of terms.
 Taking a hint from these representations, we assume that a
function f(x) having the similar properties can be expressed as
f ( x)  a0  a1 ( x  x0 )  a2 ( x  x0 ) 2  a3 ( x  x0 )3  ......  an ( x  x0 ) n  .......
Note that above three series can be obtained from this form if we
take x0 = 0 and appropriate values of coefficients on right hand side.
Differentiating both sides repeatedly with respect to x, we get
f ( x)  a1  2a2 ( x  x0 )  3a3 ( x  x0 ) 2  ......  nan ( x  x0 ) n 1  .......
f ( x)  2a2  3.2a3 ( x  x0 )  ......  n(n  1)an ( x  x0 ) n  2  .......
f ( x)  3.2a3  ......  n(n  1)(n  2)an ( x  x0 ) n 3  .......
......................................................
f (n)
( x)  n(n  1)(n  2)......3.2an  (n  1)n(n  1)(n  2)......3.2an 1 ( x  x0 )  .....
......................................................
Now taking x = x0, we obtain f ( x0 )  a0 f ( x0 )  a1 f ( x0 )  2a2

f ( x0 )  3.2a3 f ( n ) ( x0 )  n(n  1)(n  2)......3.2an

 1
Therefore a0  f ( x0 ) a1  f ( x0 ) a2  f ( x0 )
2
1 1 (n)
a3  f ( x0 ) an  f ( x0 )
3! n!

Hence the series for f(x) takes the form

( x  x0 )2
f ( x)  f ( x0 )  f ( x0 )( x  x0 )  f ( x0 ) 
2!
( x  x0 )3 ( x  x ) n
f ( x0 )  ......  f ( n ) ( x0 ) 0
 .......
3! n!
This is known as Taylor’s Series.
If we take x0 = 0, the resulting series
x2 x3 x n
f ( x)  f (0)  f (0) x  f (0)  f (0)  ......  f ( n ) (0)  .......
2! 3! n!
is known as Maclaurin Series.
Taylor’s Formula:
If we take only the terms up to and including nth derivative in the
expansion, then
( x  x0 ) 2
f ( x)  f ( x0 )  f ( x0 )( x  x0 )  f ( x0 )
2!
( x  x0 )3 ( x  x0 ) n
 f ( x0 )  ......  f ( x0 )
(n)
 Rn
3! n!
Rn the remainder after nth derivative term gives the error in the
computed approximate value of the function
Its magnitude is given by
n 1
( x  x )
Rn  f ( n1) (c{x  x0 }) 0
(n  1)!
Example:
Maclaurin series for sin x is
x3 x5 x 7
sin x  x     .............
3! 5! 7!
Polynomial approximations are
x3
P1 ( x)  x  P2 ( x) P3 ( x)  x   P4 ( x)
3!
x3
If we consider the approximation x  then error is
3!
5
x
R4 ( x)  f (5) (cx) , 0  c  1
5!
f ( x)  sin x f (5) ( x)  cos x
x5
R4 ( x)  cos(cx) , 0  c 1
5!
x5
Since cos(cx)  1 therefore R4 ( x) 
5!
Problem:
For what values of x the polynomial
x3
x
3!
will give the values of sin x such that error is not more than 3×10-3.
Solution:
The greatest error (as computed above) is given by
5
x5 x

5! 120
Therefore error will be less than or equal to 3×10-3, if
5
x
x  360  103
5
3
 3  10
120
1


x  360  10 3 5
  0.815
 Series Representation of a function of two variables
If f(x, y) is partially differentiable function, then taking a hint from
the linear approximation, we write

f ( x, y )  a0  a11 ( x  x0 )  a12 ( y  y0 )
 a21 ( x  x0 ) 2  a22 ( x  x0 )( y  y0 )  a23 ( y  y0 ) 2
 a31 ( x  x0 )3  a32 ( x  x0 ) 2 ( y  y0 )  a33 ( x  x0 )( y  y0 ) 2  a34 ( y  y0 )3
 ........................
 
where f ( x0 , y0 )  a0 f ( x0 , y0 )  a11 f ( x0 , y0 )  a12
x y
2 2 2
f ( x0 , y0 )  2a21 f ( x0 , y0 )  a22 f ( x0 , y0 )  2a23
x 2
xy y 2

3 3 3
f ( x0 , y0 )  6a31 f ( x0 , y0 )  2a32 f ( x0 , y0 )  2a33
x 3
x y
2 xy 2

3
f ( x0 , y0 )  6a34
y 3
 Therefore
f f
f ( x, y )  f ( x0 , y0 )  ( x  x0 )  ( y  y0 )
x y
1 2 f  2
f 1  2
f
 ( x  x ) 2
 ( x  x )( y  y )  ( y  y ) 2

2 x 2 xy 2 y 2
0 0 0 0

1 3 f 1 3 f
 ( x  x ) 3
 ( x  x ) 2
( y  y0 )
6 x 2 x y
3 0 2 0

1 3 f 1  3
f
 ( x  x )( y  y ) 2
 ( y  y ) 3
 ...................
2 xy 6 y
2 0 0 3 0

All partial derivatives are at (x0, y0).

 f f
f ( x, y )  f ( x0 , y0 )  ( x  x0 )  ( y  y0 )
x y
1  2 f  2
f  2
f 2
  2 ( x  x0 )  2
2
( x  x0 )( y  y0 )  2 ( y  y0 ) 
2!  x xy y 
1  3 f  3
f
  3 ( x  x0 )  3 2 ( x  x0 ) 2 ( y  y0 )
3

3!  x x y
3 f 3 f 3
3 ( x  x0 )( y  y0 )  3 ( y  y0 ) 
2

xy 2
y 
 ........................

This is known as Taylor’s Series.

Taylor’s Formula:
If we take only the terms up to and including nth derivative in the
expansion, then
 f f
f ( x, y )  f ( x0 , y0 )  ( x  x0 )  ( y  y0 )  ..............
x y
1  n f  n
f
  n ( x  x0 ) n  n n 1 ( x  x0 ) n 1 ( y  y0 )  ....
n !  x x y
n f n
 n ( y  y0 )   Rn ( x, y )
y 
is known as Taylor’s Formula.
If x0  0, y0  0
then Taylor’s series is called Maclaurin Series and has the form
f f 1  2 f 2 2 f 2 f 2 
f ( x, y )  f (0, 0)  x y  2 x 2 xy  2 y 
x y 2!  x xy y 
1  3 f 3 3 f 2 3 f 3 f 3 
  3 x 3 2 x y3 xy  3 y 
2

3!  x x y xy 2
y 
 ........................
Problem:
Use Taylor’s series at origin (Maclaurin series) to find the liner,
quadratic and cubic approximations for the function
f ( x, y)  xe y
Solution:
f ( x, y)  xe y f (0,0)  0

f x ( x, y)  e y f x (0,0)  1 f y ( x, y)  xe y f y (0,0)  0

f xx ( x, y)  0 f xx (0,0)  0 f xy ( x, y)  e y f xy (0,0)  1

f yy ( x, y)  xe y f yy (0,0)  0

f xxx ( x, y)  0 f xxx (0,0)  0 f xxy ( x, y)  0 f xxy (0,0)  0

f yyx ( x, y)  e y f yyx (0,0)  1 f yyy ( x, y)  xe y f yyy (0,0)  0

 Therefore the Taylor’s series at (0,0) is
f f 1  2 f 2 2 f 2 f 2 
f ( x, y )  f (0, 0)  x y  2 x 2 xy  2 y 
x y 2!  x xy y 
1  3 f 3 3 f 2 3 f 3 f 3 
  3 x 3 2 x y3 xy  3 y 
2

3!  x x y xy 2
y 
 ........................
1 1
f ( x, y )  x    3 xy 2   .......................
2 xy 
2! 3!
1 2
f ( x, y )  x  xy  xy  .......................
2
Therefore linear approximation is P1 ( x, y)  x

Quadratic approximation is P2 ( x, y)  x  xy
1 2
Cubic approximation is P3 ( x, y )  x  xy  xy
2