Lectures on Differentiation by Ghulam Farid 826

Values of the Trigonometric Ratios at Certain Angles:

π π π π
Angle θ 0 6 4 3 2
Value of 0o 30 o 45o 60o 90 o
T. Ratio
Sinθ 0 1 1 3 1
2 2 2
Cosθ 1 3 1 1 0
2 2 2
0 1 2 3 4
We can write the row of Sinθ as , , , , and then the row of Cosθ in the reverse
2 2 2 2 2
order.

y-axis
(0, 1)

P(Cosθ, Sinθ)
Q-II Q-I
S A
(  1, 0) T O x-axis
Q-III
C
Q-1V
(1, 0)

(0,  1)

Note:

 
 0 if θ  [0,
2
] i.e. in the 1st quadrant
Tanθ is 
 0 
if θ  [ , ] i.e. in the 2nd quadrant
 2
9.7 Increasing and Decreasing Function:

9.7.1 Increasing Function:

y-axis

y  f(x)

f(b )

f(x 2 )
f(x1 )
f(a )

O a x1 x2 b x-axis
Lectures on Differentiation by Ghulam Farid 827
A function y  f(x) is said to be an increasing function on an interval [a, b] if,
f(x1 )  f(x 2 )  x1 , x 2  [a, b] such that, x1  x 2

Theorem: 9.2
f(x) is increasing function on [a, b]  f  (x)  0  x  [a, b]
y-axis

y  f(x)

a b x-axis
O

Example: 9.25
Determine the interval in which the curve y  x 2  3 defined on the interval [1, 10] , is increasing or
decreasing?

Solution:
y  x2  3
dy
  2x   ve  x  [1, 10]
dx
dy
i.e. 0 on [1, 10]
dx
 y is an increasing function on [1, 10].

20
y  x2  3

15

10
y-axis

-5
-10 -8 -6 -4 -2 0 2 4 6 8 10
x-axis

9.7.2 Decreasing Function:

Lectures on Differentiation by Ghulam Farid 828
y-axis

y  f(x)

f(a )
f(x1 )
f(x 2 )
f(b )
O a x1 x2 b x-axis

A function y  f(x) is said to be a decreasing function on an interval [a, b] if,

f(x1 )  f(x 2 )  x1 , x 2  [a, b] such that, x1  x 2

Theorem: 9.3
f(x) is decreasing function on [a, b]  f  (x)  0  x  [a, b]

y-axis

y  f(x)

O a x1 x2 b x-axis

Example: 9.26
Determine the interval in which the curve y   x 2  1000 , defined on the interval [3, 40] , is increasing
or decreasing?
Solution:
y   x 2  1000
dy
  2x   ve  x  [3, 40]
dx
dy
i.e. 0  x  [3, 40]
dx
 y is a decreasing function on [3, 40].
Lectures on Differentiation by Ghulam Farid 829

1500

1000

500
y-axis

y  -x 2  1000
-500

-1000

-1500
-50 -40 -30 -20 -10 0 10 20 30 40 50
x-axis

Example: 9.27
Determine the interval in which the curve y  3x 2  12x  2 , defined on the interval [  8, 4] , is
increasing or decreasing?

Solution:
Given that, y  3x 2  12x  2
dy
  6x  12
dx
 ve   8  x  2
dy 
i.e. is 0 at x  2
dx  ve   2  x  4

a decreasing function on  8,  2

 y is constant at x  2
a increasing function on
  2, 4
50

40

y  3x2  12x  2
30

20
y-axis

10

-10

-20

-30
-6 -5 -4 -3 -2 -1 0 1 2
x-axis

Example: 9.28
Lectures on Differentiation by Ghulam Farid 830
The weight (in pounds) of a newborn infant during its first three months of life can be modeled by,
1 5 19
W  t3  t 2 - t  8 (where t is measured in months).
3 2 6
Determine when the infant was gaining weight and when it was losing weight?
Solution:
We are asked to find the interval where the function W is increasing and when it is decreasing.
1 5 19
We have, W  t3  t 2 - t 8
3 2 6
dW 1 2 5 19
  (3t )  (2t) -
dt 3 2 6
dW 19
i.e.  t 2  5t -
dt 6
dW 19
0  t 2  5t -  0
dt 6
i.e. 6t  30t -19  0
2

-30  (30) 2  4(6)(19)

 t
2(6)
-30  (30) 2  4(6)(19) -30  36.82 -30  36.82
i.e. t   0.568, - 5.57
2(6) 12 12
i.e. t 0.568 as - ve value of t is inadmissible.
dW
Thus, 0 at t 0.568
dt
-ve  t   0, 0.568
dW 
It is easy to see that,  0 at t  0.568
dt 
 ve  t  0.568, 3
a decreasing function on  0, 0.568

 W is constant at t  0.568
an increasing function on 0.568, 3
  
We can conclude that the infant was losing weight for the first 0.568 months of its life and then began
gaining weight afterwards at least up to the third month

30

25

1 3 5 2 19
W t  t - t 8
20
3 2 6
y-axis

15

10

0
0 0.5 1 1.5 2 2.5 3
x-axis
Lectures on Differentiation by Ghulam Farid 831

9.8 Critical Point:

A point x  c  Df is said to be a critical point for the curve y  f(x) if f (c)  0 or f is not
differentiable at x = c.
In other words, a point x  c  Df at which the tangent line becomes horizontal or vertical, is called a
critical point of the curve y  f(x) .

y-axis y-axis

O c2 x-axis O c4 x-axis

y-axis y-axis

O c1 x-axis O c3 x-axis
y-axis
y-axis

O c6 x-axis
O c8 x-axis

y-axis
y-axis

O c5 x-axis
O c7 x-axis
In the above figures, each ci is a critical point while c 2 , c 4 , c6 and c8 are stationary points.
9.8.1 Stationary Point:
A point x  c  Df is said to be a stationary point for the curve y  f(x) if f (c)  0 .
In other words a point x  c  Df at which the tangent line becomes horizontal, is called a stationary
point of the curve y  f(x) .

9.8.2 Saddle Point:

A point x  c  Df is said to be a saddle point for the curve y  f(x) , if f has no extrema at x  c .
9.8.3 Extreme Points (Relative and Global):
A function f is said to have a relative extrema at a point x  c , if it has either a relative minima or a
relative maxima at the point x  c .
Lectures on Differentiation by Ghulam Farid 832

y  f(x)
y-axis

a b
x1 O x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x-axis

In the figure, x 2 , x 4 , x 6 , x 8 , x10 are the points of relative maxima while x1 , x 3 , x 5 , x 7 , x 9 , x11 are the
points of relative minima (each of these points x1 , x 2 , x 3 , . . . . . , x11 is an extreme point). Also in the
figure, x 6 is a point of global maxima and x 3 is a point of global minima.
undefined (i.e. the tangent line becomes  to x-axis) for i  9
Note that f (x i )  
0 (i.e. the tangent line becomes to x-axis) for i  9

9.8.3.1 Absolute (Global) Maxima:

A function y  f (x) defined on [a, b] is said to have an absolute maxima on [a, b] (i.e. at a point
x  c  [a, b] ) if there is a point c  [a, b] such that, f(c)  f(x)  x  [a, b] . The value f(c) is
called the absolute maximum value of f on [a, b] .
9.8.3.2 Absolute (Global) Minima:
A function y  f (x) defined on [a, b] is said to have an absolute minima on [a, b] (i.e. at a point
x  c  [a, b] ) if there is a point c  [a, b] such that, f(c)  f(x)  x  [a, b] . The value f(c) is
called the absolute maximum value of f on [a, b] .

9.8.3.3 Relative Maxima :

A function y  f (x) defined on [a, b] is said to have a relative maxima at a point x  c  a, b if
there exists a real number δ  0 such that, f(c)  f(x)  x  [c - δ, c  δ]  [a, b] . The value
f(c) is called a relative maximum value of f at x  c .

9.8.3.4 Relative Minima :

A function y  f (x) defined on [a, b] is said to have a relative minima at a point x  c  a, b if
there exists a real number δ  0 such that, f(c)  f(x)  x  [c - δ, c  δ]  [a, b] . The value
f(c) is called a relative minimum value of f at x  c .

Theorem: 9.4
Let x  c be an extreme point for a function y  f (x) , then either f (c)  0 or f (c) is undefined.
Lectures on Differentiation by Ghulam Farid 833
That is, the relative extrema of a function y  f (x) , if any, occurs at critical points of the function.
Proof:
Case-a
Suppose that, x  c be an point of relative maxima for the function y  f (x) .
Therefore, there exists a real number δ  0 such that,
f(c)  f(x) . . . . . (1)  x  [c - δ, c  δ]  [a, b]
f(x) - f(c)
But, f (c)  lim
x c x -c
Case-a1
f(x) - f(c)
If lim does not exist, f (c) does not exist.
x c x -c
That is, f is not differentiable at x  c .
Case-a2
f(x) - f(c)
If lim exists, f (c) exists.
x c x -c
 Lf (c)  Lf (c)
i.e. f (c)  Lf (c)
& f (c)  Rf (c)
f(x) - f(c)
i.e. f (c)  lim
x  c-0 x -c
f(x) - f(c)
& f (c)  lim
x  c+0 x -c
-ve (or) 0
 f (c) 
-ve
-ve (or) 0
& f (c)  by (1)
 ve
 f (c)  0
& f (c)  0
 f (c)  0

Case-b
Suppose that, x  c be an point of relative minima for the function y  f (x) .
Therefore, there exists a real number δ  0 such that,
f(c)  f(x) . . . . . (2)  x  [c - δ, c  δ]  [a, b]
f(x) - f(c)
But, f (c)  lim
x c x -c
Case-b1
f(x) - f(c)
If lim does not exist, f (c) does not exist.
x c x -c
That is, f is not differentiable at x  c .
Case-b2
f(x) - f(c)
If lim exists, f (c) exists.
x c x -c
 Lf (c)  Lf (c)
i.e. f (c)  Lf (c)
& f (c)  Rf (c)
f(x) - f(c)
i.e. f (c)  lim
x  c-0 x -c
f(x) - f(c)
& f (c)  lim
x  c+0 x -c
Lectures on Differentiation by Ghulam Farid 834
 ve (or) 0
 f (c) 
-ve
 ve (or) 0
& f (c)  by (2)
 ve
 f (c)  0
& f (c)  0
 f (c)  0

Necessary Conditions For Extreme Point:

For the curve y  f (x) defined on an interval I (  Df ) , the necessary (but not sufficient) condition for a
 f  (c) = 0
point c  Df to be an extreme point of f is,  .
or f  (c) does not exist
  0
 x  c is an extreme point of f  f  (c)  
 or undefined
That is, 
but f  (c)   0
 x  c may not be an extreme point
 
 or undefined

9.9 Concavity:
9.9.1 Concave Up:

y-axis

(b, f(b))

(a, f(a))
y  f(x)

x-axis
O (a, 0) (b, 0)

Concave Up

The graph of a function y  f(x) is said to be concave up in the interval a, b , if it lies above every
tangent line at the points between (a, f(a)) and (b, f(b)) on the curve.
9.9.2 Concave Down:
Lectures on Differentiation by Ghulam Farid 835

y-axis

(b, f(b))

(a, f(a)) y  f(x)

x-axis
O (a, 0) (b, 0)

Concave Down

The graph of a function y  f(x) is said to be concave down in the interval a, b , if it lies below every
tangent line at the points between (a, f(a)) and (b, f(b)) on the curve.
Theorem: 9.5
If y = f(x) possesses continuous 2nd order derivatives in a, b , then
 The curve is concave up in a, b iff f  (x)  0  x  a, b

 The curve is concave down in a, b iff f  (x)  0  x  a, b

 The curve is concave up at c  a, b if f  (c)  0

 The curve is concave down at c  a, b if f  (c)  0

Note: Left and Right Neighborhood of a Point

N(c)
c
c -δ cδ
LN(c)
RN(c)

Let δ be a positive real number, then we define the interval [c - δ, c  δ] , a neighborhood of the
point c and is denoted by N(c) . Also we define the interval  c - δ, c (  LN(c)) , a left and
c, c  δ (  RN(c)) a right neighborhood of the point c. If f is a function such that,

 ve  x   c - δ, c
 -ve  x  c - δ, c

f(x)   or f(x)  
-ve  x  c, c  δ
  ve  x  c, c  δ


then we will say that f changes its sign on left and right neighborhood of c.

Maxima and Minima:

For a continuous function y = f(x) defined on [a, b],

Step-1
Lectures on Differentiation by Ghulam Farid 836
dy
 0 gives possible extreme points (i.e. stationary points) x  x1 , x 2 , x 3 (say)
dx

Step-2

 ve (definite no.) at x  x1
d2 y 
If 2
is  ve (definite no.) at x  x 2
dx 0 (or) undefined
 at x  x 3

Maximum at x  x1 
 
 Minimum at x  x 2 
then, y is  
2nd derivative test fails (we must use 1st derivative test to analyze it, 
 this x  x 3 point may or may not be an extreme point) 


Point of Relative Minima:

y-axis

y  f(x)

f =0

f <0 f >0

x-axis
O ah a ah

A function y  f(x) is said to have a relative minima at a point x  a , if there exist a neighborhood
 ve  x  LN(a)
N(a) of a such that, f  (x)   (1st der. test)
 ve  x  RN(a)

Note:
f (a)   ve  x  a is a point of relative minima for the curve.
The converse of this result may not be true.

Point of Relative Maxima:

Lectures on Differentiation by Ghulam Farid 837
y-axis

y  f(x)

f >0 f <0
f =0

x-axis
O b h b bh

A function y  f(x) is said to have a relative maxima at a point x  b , if there exist a neighborhood
 ve  x  LN(b)
N(b) of b such that, f  (x)   (1st der. test)
  ve  x  RN(b)

Note:
f (a)  -ve  x  b is a point of relative maxima for the curve.
The converse of this result may not be true.

9.10 Point of Inflection:

A point (c, f(c)) on the curve y f(x) is said to be a point of inflection if the curve is concave up on one
side of c and concave down on the other side, and is continuous at x  c .

y-axis y  f(x)
Tp

P(c, f(c))

c x-axis
O

Note:

 The tangent line at a point of inflection of a curve always crosses the curve.
 If (c, f(c)) is a point of inflection then f  (c) = 0
 If f  (c) = 0 then (c, f(c)) may or may not be point of inflection.

That is, f  (c)  0 is necessary but not sufficient condition for the point (c, f(c)) to be a point of
inflection.
Note:
f  (x)  0 x  ci (possible point of inflection)
This point x  ci will be a point of inflection if there exists a neighborhood N(ci ) such that, f (x)
changes its sign on LN(ci ) and RN(ci ) .
Lectures on Differentiation by Ghulam Farid 838
y-axis
y  f(x) f <0

f >0

f <0

f >0

x-axis
O ci  ε ci ci  ε

 ve  x  LN(ci )
That is, f  (x)  
-ve  x  RN(ci )
Note:
1st derivative test is general test to know about an extreme point, whether it is a point of relative maxima
or minima. Whereas the 2nd derivative test may fail for some cases.

Example: 9.29
Find the extreme point of the curve y  x 2 and discus its nature?
Solution:
y  x2
y  2x
y  2
For relative extrema, y  0
 2x  0
 x0
Now, y  2
 2 at x  0
i.e. y  0 at x  0
Hence, y is minimum at x  0
i.e. x  0 is the point of relative minima for the curve.
i.e. y min.  y at x 0  (0) 2  0
10

y  x2
5
y-axis

-5
-10 -8 -6 -4 -2 0 2 4 6 8 10
x-axis

Example: 9.30
2
Locate the points of relative extrema of y  3x 3 ?
Lectures on Differentiation by Ghulam Farid 839
Solution:
2
y  3x 3
2 2 1 
1
2
y  3( )x 3  2x 3  1
3
x3
1
1  1 2 4 2
y  2( )x 3   x 3   4
3 3
x3
For relative extrema y   as y  0 x
2
i.e. 1

3
x
x0
Now y is undefined at x  0
i.e. 2nd derivative test fails in this case.

We use 1st derivative test, alternatively,

 ve for x<0 (small ngd.)
Hare, y = 
 ve for x>0 (small ngd.)
Thus, x  0 is the point of relative minima for the curve.

4
2
y3x 3

2
y-axis

-1

-2
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
x-axis
Lectures on Differentiation by Ghulam Farid 840

1
Differentiate (2 x  3) 2 by definition.
2 x 1  x2
Differentiate with respect to .
1  x2 1  x2
2
Locate the points of relative extrema of y  3x .
3