Application of Partial

Differentiation

❑ Jacobians and functional dependence.

❑ Errors and Approximation.

❑ Maxima and Minima of a function of two independent variable.

❑ Lagrange’s method of undetermined multipliers.

JACOBIANS
 JACOBIANS
• Definition:
If u and v be continuous and differential function of two other
independent variable x and y such as u = 1 ( x, y ), v = 2 ( x, y ) then
we define the determinant
u u
x y
as Jocobian of u, v with respect to x,y.
v v
x y
 (u , v )
And often denote this as  ( x, y )
or sometimes also

 (u , v ) ux uy
J = =
 ( x, y ) vx vy
In the same way if u, v, w be the continuous and differential
Functions of other variable x, y, z then we can define

ux uy uz
 (u , v, w) as a Jocobian of u, v, w with
vx vy vz = respect to x, y, z.
 ( x, y , z )
wx wy wz

Also define as
ux uy uz
 (u , v, w)
J= = vx vy vz
 ( x, y , z )
wx wy wz
 Examples on Jacobians
Example 1: Compute the Jacobian of the polar coordinates
x = r cosθ, y=r sinθ.

Sol. Here, x,y are the functions of r and θ

So,  ( x, y ) xr x
J = =
 (r , ) yr y
Example 2 : Compute the Jacobian for

Sol. (Hint) xr x x
 ( x, y , z )
J = = yr y y
 (r , ,  )
zr z z
Example 3 :
Example 4:
PROPERTIES OF JACOBIANS
 (u, v) ux uy
If J = =
 ( x, y ) vx vy

Then  ( x, y ) xu xv
J = =
 (u , v ) yu yv

This implies ,

 (u , v )  ( x, y )
J  J =  =1
 ( x, y )  (u , v )
 Examples
1.

Sol.
2.

Sol.
3.
4.
Homework Examples
 Jacobians of composite
functions
• Chain Rule for Jacobians:
if x, y are the functions of u, v and u, v are the function of r, s
such that, x = 1 (u , v), y = 2 (u , v) and u = 1 (r , s ), v =  2 (r , s )
Then ,
 ( x, y )  ( x, y )  (u , v)
= 
 (r , s )  (u , v)  (r , s )
Similarly,
if x, y, z are the functions of u, v, w and u, v, w are the function of
r, s, t then,
 ( x, y, z )  ( x, y, z )  (u , v, w)
= 
 (r , s, t )  (u , v, w)  (r , s, t )
 Examples
Example.1
Example.2
 Home work
1.
 Jacobians of Implicit
functions
• Chain Rule for Jacobians:
if u1 , u2 , u3 ...un be the implicit functions of the variable x1 , x2 , x3 ...xn
connected by f1 , f 2 , f 3 ... f n such that,
f1 (u1 , u 2 ,...u n , x1, x2 ,...xn ) = 0, f 2 (u1 , u 2 ,...u n , x1, x2 ,...xn ) = 0,...
f 3 (u1 , u 2 ,...u n , x1, x2 ,...xn ) = 0
Then,
 ( f1 , f 2 ,... f n )
 (u1 , u2 ,...un )  ( x1 , x2 ,...xn )
= (−1) n

 ( x1 , x2 ,...xn )  ( f1 , f 2 ,... f n )
 (u1 , u2 ,...un )
 Remarks
• if u1 , u2 , u3 be the implicit functions of the variable x1 , x2 , x3
connected by f1 , f 2 , f 3 such that,
f1 (u1 , u2 , u3, x1, x2 , x3 ) = 0, f 2 (u1 , u2 , u3, x1, x2 , x3 ) = 0, f 3 (u1 , u2 , u3, x1, x2 , x3 ) = 0
 ( f1 , f 2 , f 3 )
Then,  (u1 , u 2 , u3 )  ( x1 , x2 , x3 )
= ( −1) 3
 ( x1 , x2 , x3 )  ( f1 , f 2 , f 3 )
 (u1 , u 2 , u3 )

• if u1 ,u2 be the implicit functions of the variable x1 , x2

connected by f1 , f 2 such that, f1 (u1 , u2 , x1, x2 ) = 0, f 2 (u1 , u2 , x1, x2 ) = 0
 ( f1 , f 2 )
 (u1 , u 2 )  ( x1 , x2 )
= ( −1) 2
 ( x1 , x2 )  ( f1 , f 2 )
 (u1 , u 2 )
Examples
 Home work
1.

2.
 FUNCTIONAL DEPENDENCE
Let u = f1 ( x, y ) and v = f 2 ( x, y ) be any two function of x, y .
Sometimes we study under what condition u = f1 ( x, y ) and
v = f 2 ( x, y ) Will be functionally dependant or independent.

If there exists a functional relation between u = f1 ( x, y) and

v = f 2 ( x, y) of the type v = F (u ) , then we say that u = f1 ( x, y )
and v = f 2 ( x, y) are functionally dependent.
For example ,
y x
if , u = , v =
x y
are Functionally Dependent
1
v=
u
 Remark
❑ Let u = f1 ( x, y) and v = f 2 ( x, y ) be any two function of x, y .
are functionally dependent if
 (u , v)  ( f1 , f 2 )
J= = =0
 ( x, y )  ( x, y )

For examples , y x
if , u = ,v =
x y
 (u , v) u x uy
J= =
 ( x, y ) v x vy
−y
x2
1
x  xy 1
= −x
= 2 2 − =0
x y xy 
1
y y2

=> u and v are Functionally Dependent.

 Examples

Sol.
Sol.
=> u and v are Functionally Dependent.
 Home work
❑.

❑.

❑.

❑.

❑.
 ERRORS AND
APPROXIMATION
Let z = f ( x, y ) ......(i)
If x, y are small increments in x and y respectively
and z is small increments in z.
then, z + z = f ( x + x, y + y ) ........(ii )
Subtracting (i) from (ii), we get,
z = f ( x + x, y + y ) − f ( x, y )
= f ( x, y ) + x fx + y fy + ...... − f ( x, y )
= x fx + y fy (Approximately)
As neglecting Higher power of x, y
 so, z = x fx + y fy (Approximately)
 If x, y are small changes (Error) in x and y respectively
then an approximate change (or Error) in z is z
Now, Replacing x, y, z by dx, dy, dz respectively.
We have
f
dz = x dx + fy dy
Here, dz , dx, dy are the Actual Error in z , x, y respectively
Note that :
dx dy dz
• , , are known as Relative Error in x, y, z respectively.
x y z
100dx 100dy 100dz
• , , are known as Percentage Error in x, y, z respectively.
x y z
 Given that,

 % Error in L = 1
% Error in G = 2

 Given that,

 % Error in E = 3
% Error in R = −2

 Given that,

% Error in x, y, z = 2
 Home work
❑.

❑.

❑.

❑.

❑.

❑.
 MAXIMA AND MINIMA
❑Maxima and Minima of Functions of y=f(x) :-
• The problem of determining the maximum or minimum of
function is encountered in geometry, mechanics, physics, and
other fields, and was one of the motivating factors in the
development of the calculus in the seventeenth century.
• Let us recall the procedure for the case of a function of one
variable y=f(x). First, we determine points “a” from
f'(x)=0(solving this equation). These points are called critical
points / stationary points. At critical points the tangent line is
horizontal. This is shown in the figure below.
• The second derivative test is employed to determine if a
critical point / stationary points is a relative maximum or a
relative minimum.
• If f''(a)>0, then “a” is a relative minimum.
• If f''(a)<0, then “a” is a maximum.
• If f''(a)=0, then the test gives no information.
• For example:
 3 x 2 − 3 = 0  3( x 2 − 1) = 0
x 2 − 1 = 0  ( x − 1)( x + 1) = 0
x = 1 / x = −1
Now we have to determine whether any of
these stationary points are extremum points.
We’ ll use the second derivative test for this :
dy d2y
= 3x − 3 
2
= 6x
dx dx 2
❑ Maxima and Minima of Functions of z=f(x,y):-

❑ Critical / Stationary Points :-

• The notions of critical points and the second derivative test carry
over to functions of two variables. Let z=f(x,y). Critical points are
points in the xy-plane where the tangent plane is horizontal.

• Hence, critical points are solutions of the equations:

f x(x,y) = 0 and f y(x,y) = 0

• Lets find the critical points for
 1 3 2
 − x + x− y 
z = f ( x, y ) = e  3 

the partial derivatives are

 1 3 2
 − x + x− y 
f x ( x, y ) = (− x 2 + 1)  e  3 

 1 3 2
 − x + x− y 
f y ( x, y ) = (−2 y )  e  3 

now, f x ( x, y ) = f y ( x, y ) = 0
gives us
(− x 2 + 1) = 0 and (−2 y ) = 0
 x = 1, y = 0
critical / stationary po int s are
(1,0) and (-1,0)
• Notice the relative maximum at (x=1,y=0). (x=-1,y=0) is a
relative maximum if one travels in the y direction and a
relative minimum if one travels in the x-direction. Near (-1,0)
the surface looks like a saddle(neither maxima nor minima).

 1 3 2
 − x + x− y 
z = f ( x, y ) = e  3 
❑ The Second Derivative Test for Functions of
Two Variables:-
• How can we determine if the critical points found above are
relative Maxima or Minima?
• We apply a second derivative test for functions of two
variables.
Let (a,b) be a critical point and define rt − s 2
where,
 2 f   2 f   2 f 
r =  2  , t =  2  , s =  
 x  ( a ,b )  y  ( a ,b )  xy  ( a ,b )
or
r = f xx (a, b), t = f yy (a, b), s = f xy (a, b)
now
• if rt − s 2  0, and r = f xx (a, b)  0  f ( x, y ) is Minimum at (a, b)
and f (a, b) its Minimum value.
• if rt − s 2  0, and r = f xx (a, b)  0  f ( x, y ) is Maximum at (a, b)
and f (a, b) its Maximum value.
• if rt − s 2  0  f ( x, y ) is neigther Maxima nor Minimum at (a, b)
( Such a point is called as saddle po int )
• if rt − s 2 = 0  the case is undecided
  1 3 2
 − x + x− y 
Lets continued previous example z = f ( x, y ) = e  3 

( ) ) e
 1 3 2  1 3 2

( ) (
 − x + x− y   − x + x− y 
2 2
fx = − x +1  e
2  3 
 f xx ( x, y ) = − 2 x + 1 − x  3 

 1 3 2  1 3 2

( )
 − x + x− y   − x + x− y 
f y = (− 2 y )  e  3 
 f yy ( x, y ) = − 2 + 4 y 2  e  3 

 1 3 2

( )
 − x + x− y 
f xy ( x, y ) = −2 y 1 − x 2  e  3 

• Now, for po int (1,0)

( 23 ) ( 23 )
r = f xx (1,0) = −2  e , s = f xy (1,0) = 0, t = f yy (1,0) = −2  e
( 43 ) ( 23 )
 rt − s = 4e
2
 0 and r = f xx (1,0) = −2  e 0
 (1,0) is Maximum Po int .
  1 3 2
 − x + x− y 
Lets continued previous example z = f ( x, y ) = e  3 

( ) ) e
 1 3 2  1 3 2

( ) (
 − x + x− y   − x + x− y 
2 2
fx = − x +1  e
2  3 
 f xx ( x, y ) = − 2 x + 1 − x  3 

 1 3 2  1 3 2

( )
 − x + x− y   − x + x− y 
f y = (− 2 y )  e  3 
 f yy ( x, y ) = − 2 + 4 y 2  e  3 

 1 3 2

( )
 − x + x− y 
f xy ( x, y ) = −2 y 1 − x 2  e  3 

• Now, for po int (−1,0)

( −32 ) ( −32 )
r = f xx (−1,0) = 2  e , s = f xy (−1,0) = 0, t = f yy (−1,0) = −2  e
(− 43 )
 rt − s = −4e  0
2

 (−1,0) is Saddle Po int .

• Notice the relative maximum at (x=1,y=0). (x=-1,y=0) is a
relative maximum if one travels in the y direction and a
relative minimum if one travels in the x-direction. Near (-1,0)
the surface looks like a saddle(neither maxima nor minima).

 1 3 2
 − x + x− y 
z = f ( x, y ) = e  3 
Revision : -
Working rule to find Extremum Values for a function z = f(x, y) :
 Examples
Example 1 : Find the Maximum and Minimum of the function
f ( x, y ) = x 3 + y 3 − 3axy
f ( x, y ) = x 3 + y 3 − 3xy
f ( z) = y 2 − x 2
Example 3 : Test the function f ( x, y ) = x 3 y 2 (6 − x − y )
for Maxima and minima for points not at the origin
 Example 4. Examine for minimum and maximum value for
f ( x, y ) = sin x + sin y + sin( x + y )

f ( x, y ) = sin x + sin y + sin( x + y )

 Homework Examples

−( x2 + y 2 )
2) test the function f ( x, y ) = ( x + y )e
2 2
for Maxima or Minima for
points not on the circle x 2 + y 2 = 1

3. In plane triangle ABC find the maximum value of

cosA  cosB  cosC
 LAGRANGE’S METHOD OF
UNDETERMINED MULTIPLIERS

We know how to find the critical points of a function of two

variables:
f f
look for where ∇f = 0. That is, = =0
x y

Sometimes, however, we have a constraint which restricts us from

choosing variables freely
❑.Maximize volume subject to limited material costs
❑. Minimize surface area subject to fixed volume
❑.Maximize utility subject to limited income
Consider the Example :
 LAGRANGE’S METHOD OF
UNDETERMINED MULTIPLIERS
❑ In many problems, a function of two or more variables is to be
optimized, subjected to a restriction or constraint on the
variables, here we will consider a function of three variables to
study Lagrange's method of undetermined multipliers

❑ Suppose it is required to find the stationary values for the

function f(x , y , z) subject to condition Φ( x , y , z) = 0 -------- (1)
❑ .Step 1 : Form a lagrangean function
F(x , y , z) = f(x , y , z) + λ Φ ( x , y , z) where “λ ” is called Lagrange’s
constant, which is determined by the following conditions.
 F f 
=0 + = 0..................(2)
x x x
F f 
=0 + = 0..................(3)
y y y
F f 
=0 + = 0..................(4)
z z z
❑. Step 2 : Solving the equations (1) (2) (3) & (4) we get the
stationary point (x, y, z).
❑. Step 3 : Substitute the value of x , y , z so obtained in
equation (1) we get the stationary Values.
 Examples
Example 1 : As dimension of a triangle ABC are varied, Show that
the maximum value of cosA  cosB  cosC is obtained when the
triangle is equilateral
Example 5 : Use Lagranges method to find the minimum distance from
origin to the plane 3 x + 2 y + z = 12
Example 6 : Divide 24 into three parts such that continued product
of the first, square of second, and cube of third may be maximum
 Home Work Examples
Example 1. FInd the Maximum Value of u = x 2 y 3 z 4
9
a
such that 2 x + 3 y + 4 z = a Answer :  
9
Example 2. Find the Maximum & Minimum value of
x 2 + y 2 when 3 x 2 + 4 xy + 6 y 2 = 140
Answer : Max value : 70, Min Value : 20

Example 3. Find Stationary values of u = x + y + z

if xy + yz + zx = 12 Ans : 6