Lecture Notes for math111: Calculus I.

Dr. Vitaly A. Shneidman

New Jersey Institute of Technology

Fall, 2013

1
 Introduction

y = f (x)

Limits & Continuity

Rates of change and tangents to curves

2
Average rate of change

∆y f (x 2 ) − f (x 1 )
= (1)
∆x x2 − x1

Equation for the secant line:

∆y
Y = (x − x 1 ) + f (x 1 )
∆x
Tangent - secant in the limit x2 → x1 (Or, ∆x → 0).
Slope of tangent
∆y
f ′ (x1) ≡ , ∆x → 0 (2)
∆x
(notation f ′ is not yet in the book, but will appear
later...). Equation of tangent line:
Y = f ′ (x 1 ) ( x − x 1 ) + f (x 1 )
3
Example (Galileo): t instead of x or x1, h instead of
∆x.
1 2
y(t) = at
2
where a is a constant (”acceleration”). Secant (finite
h)
1 h 2 2
i 1  2

∆y = y(t+h)−y(t) = a (t + h) − t = a 2th + h
2 2
∆y ∆y
= = a(t + h/2)
∆t h
Tangent (h → 0):

y ′(t) = at
(known as ”instantaneous velocity” or ”instantaneous
4
rate of change”). Returning to math notations
′
2

x = 2x (3)
Example: y(x) = x3

Use a3 −b3 = (a−b)(a2 +ab+b2) with a = x+h, b = x.

Slope of secant (A.R.C.), finite h ≡ ∆x:
3 3 2 2
h i
∆y = y(x+h)−y(x) = (x+h) −x = h (x + h) + x(x + h) + x =

= h(3x2 + 3xh + h2)

∆y ∆y
= = 3x2 + 3xh + h2
∆x h
Slope of tangent (I.R.C.) h → 0:
′
3 = 3x2

x (4)

5
 √
Example: y(x) = x
√ √
∆y ∆y x+h− x
A.R.C. =
= =
∆x h h
√ √ √ √
x+h− x x+h+ x 1
= ·√ √ =√ √
h x+h+ x x+h+ x
Now, no problem h → 0:
√
I.R.C. = 1/2 x
or
√  ′ 1
x = √ (5)
2 x

6
Continuity and Limits
1.0 1.0

0.5 0.5

1.5 2.0 2.5 3.0 1.5 2.0 2.5 3.0

-0.5 -0.5

-1.0
-1.0

Continuous (left) and discontinuous (right) functions

Intermediate value Theorem 11, p.99

7
Limit of a function
Assume don’t know f (x0) but know f (x) for any x
close to x0 . Example:
sinHxL
x

1.0

0.8

0.6

0.4

0.2

-15 -10 -5 5 10 15

-0.2

y(0) not defined. Can define y(0) = a, with any a.

Why a = 1 is the ”best”?
8
Limit:
lim f (x) = L
x→x0
Example:
x2 − a 2
lim = lim x + a = 2a
x→a x − a x→a
Functions with no limit as x → 0:
step-function, 1/x, sin(1/x)

9
Limit Laws - Theorem 1, p.68.
If f , g have limits L & M as x → c. Then,

lim (f ± g) = L ± M , lim (f · g) = L · M , etc. (6)

x→c x→c
lim (f n) = Ln , lim (f 1/n) = L1/n (7)
x→c x→c

”Dull functions” limx→c f (x) = f (c). Example

q √
lim 4x2 − 3 = 13
x→−2
Example: any polynomial limx→c Pn(x) = Pn(c)

10
Limits of rational functions P (x)/Q(x) as x → c.
If Q(c) 6= 0 - ”dull”, limit R(c)/Q(c). If Q(c) = 0
check P (c): if P (c) 6= 0 - no limit; If P (c) = 0 try to
simplify.
Example:
x3 + x − 2 (x − 1)(x2 + x + 2)
lim 2
= lim =4
x→1 x −x x→1 x(x − 1)

Other functions with limits of type ”0/0”. Example:

q q q
x2 + a 2 − a x2 + a 2 − a x2 + a 2 + a
lim = lim ·q =
x→0 x2 x→0 x 2 2 2
x +a +a
q
= limx→0 1/( x2 + a2 + a) = 1/2a

11
”Sandwich theorem” : Let f (x) ≤ g(x) and
lim f (x) = lim g(x) = L .
x→c x→c
IF
f (x) ≤ u(x) ≤ g(x)
THEN
lim u(x) = L
x→c

12
Limit (sin x)/x as x → 0
B
6 AOC = x (in radians !)
C
1 sin x
S△OAC = 2

O x
Ssector OAC = 2
A

1 tan x
S△OAB = 2

Thus,

π
sin x < x < tan x , 0 < x < (8)
2
Divide by sin x and take the inverse, changing ”<” to
”>”: 1 > sinx
x > cos x. Since cos 0 = 1, from ”S.T.”

sin x
lim =1 (9)
x→0 x
13
Examples (discussed in class):

tan x
lim =1 (10)
x→0 x
sin kx
lim =k (11)
x→0 x
1 − cos x
lim =0 (12)
x→0 x
1 − cos x 1
lim = (13)
x→0 x2 2

tan x − sin x 1
lim =
x→0 x3 2

14
One-sided limits (skip definition on p. 87)
x 1 - x2
 x¤

1.0
1.0

0.8
0.5
0.6

x
-1.5 -1.0 -0.5 0.5 1.0 1.5 0.4

-0.5 0.2

x
-1.0 -1.0 -0.5 0.0 0.5 1.0

Functions with 1-sided limits.

x x
lim = 1 , lim = −1 , f (0) =?
x→0 |x|
+ x→0 |x|
−

q q
lim 1 − x2 = 0 , lim 1 − x2 = 0
x→−1+ x→1−

15
Theorem 6, p.86:
Limit of f (x) for x → c exists if and only if L.H. and
R.H. limits at x → c− , c+ exist and have the same
value (in which case, this also will be the value of the
limit of f ) .

16
 1 1
x sin x sin
x x

0.2
0.3

0.1 0.2

0.1
x
-0.4 -0.2 0.2 0.4 x
-0.05 0.05 0.10 0.15 0.20
-0.1
-0.1
-0.2

-0.2 -0.3

Tricky functions with 2- or 1-sided limits.

1 √ 1
lim x sin = 0 , lim x sin = 0
x→0 x x→0+ x

17
Continuity

at a point x = c (p. 93, or Continuity Test, p. 94):

f (c) = lim f (x)

x→c
for an interior point (and similarly for an endpoint with
corresponding one-sided limit).

Right-continuous at a point x = c :

f (c) = lim f (x)

x→c+
Left-continuous at a point x = c :

f (c) = lim f (x)

x→c−

18
 dxt

x
-2 -1 1 2 3 4 5
-1

-2

right-continuous integer ”floor” function

19
Continuous Function (CF)

CF on an interval
CF

Theorem 8, p. 95

f ± g , etc.
Inverse of a CF is a CF: e.g. ex and ln x
q
Composits of CF are CFs: e.g. f (x) , f (x) = 1 − x2
Theorems 9, 10 on p. 97

Continuous extension to a point: e.g. F (x) = sin x /x , x 6=

1; F (0) = 1
20
Limits involving ∞
x → ±∞ (skip p. 104)

2 x2 + 5 x - 7
x2 + 1

x
-10 -5 5 10

-2

-4

-6

Limits of a rational function with the same power of

denom. and numerator.
Horizontal asymptote(s) - lim(s) of f (x) as x → ±∞
21
Non-algebraic functions
x3 x 1
, e , x sin , etc.
|x|3 + 1 x
1
x sin
x

1.0

0.8

0.6

0.4

0.2

x
-4 -2 2 4
-0.2
Oblique and vertical asymptotes

x2 + x + 2
x-1

20

10

x
-4 -2 2 4

-10

22
 Derivatives

f (x + h) − f (x)
f ′(x) = lim = (14)
h→0 h
f (z) − f (x)
= lim (15)
z→x z−x

If the limit exists, this is derivative at a point x,

a.k.a:

• slope of the graph at x

• slope of the tangent to the graph at x

23
• rate of change of f with respect to x
Derivative as a function

x is not ”frozen” anymore; can consider a new func-

tion
g(x) = f ′(x)
Other notations
dy df d
f ′(x) = y ′ = = = f (x) = . . .
dx dx dx
df 

g(a) =
dx x=a

24
Examples:
d 1 1/z − 1/x x−z 1
= lim = lim =− 2
dx x z→x z−x z→x zx(z − x) x
or

d 1 1
=− 2 (16)
dx x x

d 1 1/z 2 − 1/x2 x2 − z 2 1
= lim = lim = −2
dx x2 z→x z−x z→x z 2 x2 (z − x) x3

d x d 1 1
 
= 1+ =0−
dx x − 1 dx x−1 (x − 1)2

25
One-sided derivatives
From right:
f (x + h) − f (x) f (z) − f (x)
f ′(x) = lim = lim
h→0+ h z→x+ z−x
From left:
′ f (x + h) − f (x) f (z) − f (x)
f (x) = lim = lim
h→0 − h z→x− z−x
Example: |x|′ = 1 for x > 0, = −1 for x < 0 and is
not defined at x = 0 where only r.-h. or l.-h. deriva-
tives exist. (This was a ”corner”;qother cases with no
derivative may include ”cusp” as |x|, vertical tangent
as x1/3 and a discontinuity of the original function as
|x|/x. Graphics in class.)

26
Theorem 3.1. If f (x) has a derivative at x = c it is
continuous at x = c.

Consider identity
f (c + h) − f (c)
f (c + h) = f (c) + ·h
h
and take the limit of both sides as h → 0.
Note: Converse is false, e.g. |x|.

27
Differentiation rules

(af (x) + bg(x))′ = af ′ + bg ′ (17)

Proof. Consider
af (z) + bg(z) − af (x) − bg(x) a[f (z) − f (x)] + b[g(z) − g(x)]
≡
z−x z−x
and take the limit z → x.


′ df (z) 
(f (ax)) = a  (18)
dz  z=ax

Example: (sin(ωx))′ = ω cos(ωx)

28
 [f (x) · g(x)]′ = f ′g + f g ′ (19)

Proof. Consider
f (z)g(z) − f (x)g(x) [f (z) − f (x)]g(z) + f (x)[g(z) − g(x)]
≡
z−x z−x
and take the limit z → x.

Example:

(x · x)′ = 1 · x + x · 1 = 2x

29
 !′
1 f′
=− 2 (20)
f f

Proof. Consider
1/f (z) − 1/f (x) f (x) − f (z)
≡
z−x f (z)f (x)(z − x)
and take the limit z → x.

Example: (1/x)′ = −1/x2

!′
f f ′ · g − g′ · f
= (21)
g g2
30
Derivatives of elementary functions

(xn)′ = nxn−1 , any n (22)

(sin x)′ = cos x , (cos x)′ = − sin x (23)

′
x ′ x x ′ x ln a = ax · ln a

(e ) = e , (a ) = e (24)

(ln x)′ = 1/x (25)

31
Trigonometric functions
Earlier: (sin x)′ = cos x , (cos x)′ = − sin x and

(tan x)′ = 1/ cos2 x (26)

Other basic examples: 1/ cos x , 1/ sin x , cot x , . . .

Simple harmonic motion:

dy
y(t) = A sin(ωt) , = Aω cos(ωt) (27)
dt
d2 y 2 sin(ωt) ≡ −ω 2 y(t)
= −Aω (28)
dt2

32
Chain rule

′ df dg
(f (g(x))) = · ≡ (f (g))′ · (g(x))′ (29)
dg dx

”Proof”. Let u = g(z) , v = g(x)

f (u) − f (v) f (u) − f (v) u − v df dg
(f (g(x)))′ = lim
z→x
= lim
z→x
· = ·
z−x u−v z−x dg dx

Simple example: (f (ax))′ = af ′(ax)

33
 n ′ n−1 du
(u(x) ) = nu (30)
dx
′ du
(sin u(x)) = [cos u(x)] (31)
dx
e u ′ = eu du (32)
( )
dx
1 du
(ln u)′ = (33)
u dx

′
1
q
x2 + 1 = q · (2x)
2 x2 + 1

34
Example
 ′
e −x2 = −2xe −x2

1.0

0.5

expI-x2 M

-2 x expI-x2 M
-3 -2 -1 1 2 3

-0.5

35
Example:
′
sin(x ) = 2x cos(x2)
2


10

sinIx2 M
1 2 3 4 5 2 x cosIx2 M

-5

-10

36
Example
 ′
e −e−x =e −e −x  −x 
· e

1.0

0.8

0.6
expH-ã-x L
expH-x - ã-x L
0.4

0.2

-2 -1 0 1 2 3 4 5

37
Implicit differentiation
Let F (x, y) = 0 where F is a simple known function
(e.g. F = x − y 2 describing a parabola). Then

′ ′ dy
Fx + Fy =0 (34)
dx

and
dy
= −Fx′ /Fy′
dx

38
Tangent at (x0 , y0)

dy 

Y = y0 + (x − x 0 ) (35)
dx x=x0


Normal:
!−1
dy 

Y = y0 − (x − x0) (36)
dx x=x0


39
Parabola:
dy dy 1
y 2 = x , 2y =1, =
dx dx 2y
Tangent and normal:
1
Y = y0 + (x − x0) , Y = y0 − (2y0)(x − x0)
2y0
1.5

1.0

0.5

0.0
x - y2 ‡ 0

-0.5

-1.0

-1.5
0.0 0.5 1.0 1.5 2.0 2.5 3.0

40
Ellipse:
dy dy x b 2
(x/a)2+(y/b)2 = 1 , 2x/a2+2y/b2 =0, =− 2
dx dx ya
Tangent and normal:
x0 b 2 y0 a 2
Y = y0 − (x − x0) , Y = y0 + (x − x0)
y0 a 2 x0 b 2
3

0 x2 y2
9
+ 4
‡1

-1

-2

-3
-4 -2 0 2 4

41
Hyperbola:
2 2 dy dy x
x − y = 1 , 2x − 2y =0, =
dx dx y
Tangent and normal:
x0 y0
Y = y0 + (x − x0) , Y = y0 − (x − x0)
y0 x0
3

0
x2 - y2 ‡ 1

-1

-2

-3
-3 -2 -1 0 1 2 3

42
Inverse Functions

Consider y = f (x) and solve for x = F (y) (select only

one branch, if several). Then, swap y and x. Then,

y = F (x) , with (f (F (x))) = x

is the inverse function (F ≡ f −1 in textbook).

2 √
Examples: f = x , F = x, f = ex , F = ln x, f =
sin x , F = arcsin x, etc.

Note: range of f becomes the domain of F , but the

domain of f becomes range of F for monotonic f only.
Examples: ex (monotonic) and x2 or sin x (non-monotonic).

43
Note (f (F (x))) = x for all x in the domain of F , while
F (f (x)) = x in the entire domain of f for monotonic
f only. E.g.:

eln x = x for 0 < x < ∞ , and ln (ex) = x for −∞ < x < ∞

sin(arcsin x) = x for − 1 ≤ x ≤ 1 , but

arcsin(sin x) 6= x , for |x| > π/2

44
Primitive example: f = x − 1 , F = x + 1
4

x-1
-3 -2 -1 1 2 3 x+1

-2

-4

45
Primitive example: f = x/2 + 1 , F = 2x − 2
4

x
+1
f (F (x)) = 1
2 · (2x − 2) + 1 = x
2
-1 1 2 3 2x-2

-2

-4

46
 2 √
Example: f = x , F = x
4

1
3 (f )′ = 2x , (F )′ = 2√ x

2
x2 Consider (x0 = 2 , y0 = 4) ,
x
(X0 = 4, Y0 = 2):
1
(f )′ = 4 , (F )′ = 1
4

1 2 3 4

47
 √
Example: f = x2 − 2 , F = x+2
6

(f )′ = 2x , (F )′ = √1
4 2 x+2

x2 - 2 Consider (x0 = 3 , y0 = 7) ,
2
x+2
(X0 = 7, Y0 = 3):
(f )′ = 6 , (F )′ = 1
6
-2 2 4 6

-2

48
Example: f = ex , F = ln x
6

expHxL
2
logHxL

-2 2 4 6

-2

49
Logarithm
y = ln x

ey = x

ey · y ′ = 1
1 1
y′ = y =
e x

50
Example: f = sin x , F = arcsin x
2

sinHxL
-3 -2 -1 1 2 3 sin-1 HxL

-1

-2

51
arcsin
y = arcsin x

sin y = x

(cos y) · y ′ = 1
1
y′ = =
cos y
1 1
=q =q
1 − sin2 y 1 − x2

52
Example: f = tan x , F = arctan x
4

tanHxL

-5 5 tan-1 HxL

-2

-4

53
arctan
y = arctan x

tan y = x

(1/ cos2 y) · y ′ = 1

1
y′ = 2
=
1/ cos y

1 1
= cos2 y = = 2
tan2 y + 1 x +1

54
General
y = F (x)

f (y) = x

df
· y′ = 1
dy


′ dF 1 
y ≡ = df 
dx 
dy  y=F (x)

55
Logarithmic functions and differentiation
1
(ln u(x))′ = · (u )′
u

Examples

′ b 1
(ln bx) = =
bx x

1 1
(ln |x|)′ = · (|x|)′ = , x 6= 0
|x| x

56
Recall:

x x ln a ln x
a =e , loga x = , a > 0 , a 6= 1 (37)
ln a

′
x ′ x ln a = ax · ln a

(a ) = e

1
(loga x)′ =
x ln a

n ′ n ln x
′ n n ln x
= nxn−1

(x ) = e = e
x

57
 ′ x
 
(xx)′ = ex ln x x ln x = xx(ln x + 1)

=e ln x +
x

Remarkable limit:

lim (1 + x)1/x = e (38)

x→0

58
Logarithmic differentiation

Let
P (x)Q(x)
y=
R(x)

ln y = ln P + ln Q − ln R

P ′ Q ′ R ′
( ) ( ) ( )
y ′/y = + −
P Q R

(P ) ′ (Q)′ ′
" #
′ P (x)Q(x) (R )
y = · + −
R(x) P Q R

59
Inverse trigonometric functions

′ 1
(arcsin x) = q , |x| < 1 (39)
1 − x2
1
(arctan x)′ = 2
, −∞ < x < ∞ (40)
x +1

Now, arccos x = π2 −arcsin x and cot−1 x = π2 −arctan x.

Thus

1
(arccos x)′ = − q , |x| < 1 (41)
1 − x2

−1
′ 1
cot x = − 2 , −∞ < x < ∞ (42)
x +1

60
 1 1
sec−1(x) = arccos , csc−1(x) = arcsin
x x
′ 1 ′ 1 −1
 
sec−1(x)

= arccos =− q · 2
x 1 − (1/x)2 x

1
= q , |x| > 1
|x| x2 − 1
′ 1 ′ 1 −1
 
csc−1(x)

= arcsin =q · 2
x 1 − (1/x)2 x

1
=− q , |x| > 1
|x| x2 − 1

61
Related rates

Suggested notations:
V - volume, A - area, L - length/distance, r , R - radii,
y Y, h, H - vertical position, x, X -horizontal position,
v, V velocity, t - time

Sphere:

4
V = πR3 , A = 4πR2 (43)
3

Cone:
1
V= Abase · h (44)
3

62
 1.0

Circle:

0.5

A = πr 2
0.0

dA/dt = 2πr · dr/dt:

-0.5

-1.0

-1.0 -0.5 0.0 0.5 1.0

63
 1.0

Sphere:

0.5

4 πr 3
V =3
0.0

dV = 4πr 2 · dr = A · dr
dt dt dt
-0.5

-1.0

-1.0 -0.5 0.0 0.5 1.0

64
Cone:

1 πr 2 (h) · h =
V =3

2
1 h V0

= 3 π R(H) H · h = H 3
3 ·h

dV = V0 · 3h2 dh = A(h) · dh
dt H3 dt dt

65
 ”Coffee maker” :

V1 + V2 = const , dV
dt
1 + dV2 = 0
dt

dV1 2 (h) · dh
dt = πr dt
dV2 2 · dY
dt = πR dt

dY 1 2 dh
= − 2 r (h) ·
dt R dt

66
An a(t) × b(t) rectangle:
q
A = ab , P = 2(a + b) , D = a2 + b 2

dA db da
=a +b
dt dt dt
dP da db
 
=2 +
dt dt dt
dD 1 d  2 1 da db
 
2

= a +b = a +b
dt 2D dt D dt dt

67
Distance between two moving points:

HX,YL q
L= (x − X)2 + (y − Y )2
L
Hx,yL

dL 1 d  2 2

= (x − X) + (y − Y ) =
dt 2L dt
1 dx dX dy dY
 
= (x − X)( − ) + (y − Y )( − )
L dt dt dt dt

68
Police car:

X ≡ 0, dY /dt = −V , y ≡ 0, dx/dt = v

dL 1 d  2 2

= x +Y =
dt 2L dt
1 dx dY 1
 
= x +Y = (xv − Y V )
L dt dt L
1
v= (L · dL/dt + Y V )
x

69
 q
Angle with horizontal (X, Y = 0, L = x2 + y 2 =
x/ cos θ):
1 dθ 1 dy dx
 
tan θ = y/x , = 2 x −y
cos2 θ dt x dt dt
dθ 1 dy dx
 
= 2 x −y
dt L dt dt

Balloon: x = const , dy/dt = V .

Aircraft: y = const , dx/dt = −v.

70
 y
4

2 x2
1 + 2 Hx - 1L
1

x
0.5 1.0 1.5 2.0

-1

Tangent line and error (for a parabola):

Y (x) = y(a) + y ′(a)(x − a)

y(x) − Y (x) = x2 − a2 − 2a(x − a) = (x − a)2

71
 1
2.0

1.5

x
1.0 x-1
1+ 2

0.5

x
0.0 0.5 1.0 1.5 2.0

72
 0
2

1
sinHxL
0 + 1 Hx - 0L
x
-2 -1 1 2 HsinHxL - 0 - 1 Hx - 0LL 10

-1

-2

73
Differential:

dx - independent variable

dy = y ′(x) dx

74
 1

3.0

2.5

2.0
Dy
1.5 dy
1.0
dx
0.5

x
0.6 0.8 1.0 1.2 1.4 1.6 1.8

75
 f (x + dx) ≃ f (x) + df = f (x) + f ′(x) · dx
Examples (small x instead of dx; approximation only
near x = 0):

(1 + x)2 ≃ 1 + 2x (45)
1/(1 + x) ≃ 1 − x (46)
p
1 + x ≃ 1 + x/2 (47)
(1 + x)k ≃ 1 + kx (48)
ex ≃ 1 + x (49)
ln(1 + x) ≃ x (50)

76
 y

x - sinH2 xL
4
1 - 2 cosH2 xL

x
2 4 6 8

77
Absolute maximum:

f (c) = M , f (x) ≤ M
for every x in the domain of f .

Theorem. On a closed domain [a, b] any continuous

f (x) will have an absolute maximum.

78
Local maximum:

f (c) = M , f (x) ≤ M
for every x on some open interval containing c.

Theorem. For any differentiable f (x) a local maxi-

mum will have a zero derivative:

f (c) = M , (f (x))′x=c = 0 (51)

(but, the reverse can be false: e.g. x3)

79
Critical and end points:

• endpoint(s)

• zero derivative

q
• no derivative (e.g |x| or |x|)

80
E.g.: critical point - no extrema:
y
1.0

0.5

x3
x 3 x2
-1.0 -0.5 0.5 1.0

-0.5

-1.0

81
E.g.: critical point - no extrema:
y
1.0

0.5

 x¤ sgnHxL
3

x sgnHxL Abs¢ HxL

 x¤ sgn¢ HxL
3

-1.0 -0.5 0.5 1.0 +

3  x¤ 23

-0.5

-1.0

82
 y
4

3 x2 x3
2 1+2x- +
2 3

2 - 3 x + x2
1

x
-1 1 2 3

-1

83
 y
1.0

0.5

ã-x x
x ã-x - ã-x x
-1 1 2 3

-0.5

-1.0

84
 y
2.0

1.5

1.0
ã-x x3
0.5 3 ã-x x2 - ã-x x3

x
-1 1 2 3 4 5 6

-0.5

-1.0

85
 y
1.0

0.5

x logHxL
x 1 + logHxL
0.5 1.0 1.5 2.0

-0.5

-1.0

86
 y
6

2
4 cosHxL + 3 sinHxL
x 3 cosHxL - 4 sinHxL
1 2 3 4 5 6

-2

-4

-6

87
Rolle’s Theorem:
y
0.20

0.15

0.10

0.05

x
-0.5 0.5 1.0

88
Mean Value Theorem:
y
0.20

0.15

0.10

0.05

x
-0.5 0.5 1.0

89
Corollary: If
(f )′ ≡ 0 , f ≡ const

Corollary: If

(f )′ ≡ (g ) ′ , f ≡ g + C

E.g.:
1 2
a = const , v = at + V0 , x = at + V0t + X0
2

90
MONOTONICITY

Corollary:
If (f )′ keeps sign on (a, b), f is monotonic.
y
1.0

0.5
x3
-x
x x3 - x
-2 -1 1 2
-1 + 3 x2
-0.5

-1.0

91
1st Derivative Test

• (f )′ changes from ”+” to ”-” - local max.

• (f )′ changes from ”-” to ”+” - local min.

• if (f )′ does not change sign - no local max or min

Note: Often can use 2nd derivative, but the 1st test
does not require the existence of derivatives at x = c,
e.g f = |x|.

92
 y
12

10

6 10 -  -10 + x¤ - sinHx H2 + sgnH-10 + xLLL

x
5 10 15

93
Concavity:
y
2.0

c. up1.5 c. up c. down
x2
1.0
2 - H-3 + xL2

0.5

x
-2 -1 1 2 3 4 5

94
Concave up: (f )′ increasing
Concave down: (f )′ decreasing
Change of concavity: inflection point
y
1.0

0.5
x3
-x
x x3 - x
-2 -1 1 2
-1 + 3 x2
-0.5

-1.0

95
2nd Derivative Test for Concavity:
Concave up: f ′′ > 0
Concave down: f ′′ < 0
Change of concavity: inflection point f ′′ = 0 or does
not exist

2nd Derivative Test for Local Extrema:

if f ′ = 0 and f ′′ > 0 - local min (e.g., x2)
if f ′ = 0 and f ′′ < 0 - local max (e.g., −x2)
if f ′ = 0 and f ′′ = 0 - test fails (e.g., x3 or x4)

96
 y
0.6

0.4 x4
-x2
0.2 x4 - x2
-2 x + 4 x3
x -2 + 12 x2
-2 -1 1 2

-0.2

97
 y
6

4 1
x

2 x
1
x
+x
x
-4 -2 2 4 1
1-
x2
-2
2
x3
-4

-6

98
 y
6

1
4
x2

2 x
1
+x
x2
x
-4 -2 2 4 2
1-
x3
-2
6
x4
-4

-6

99
 y

2 x
-
1+x2

x
1
x
- +x
1+x2

x x2 I3+x2 M
-4 -2 2 4 2
I1+x2 M

-1 2 x I-3+x2 M
- 3
I1+x2 M

-2

100
 y
1.0
0.8
0.6
 sinHxL¤
0.4
0.2
x
-4 -2 0 2 4

Domain:
f (x) : (−∞, ∞)

f ′(x) : (−∞, ∞) , x 6= nπ

101
 y
1.0

0.5

x 1 - x2
-1.0 -0.5 0.5 1.0
x
-
1-x2

1
- 32
I1-x2 M
-0.5

-1.0

-1.5

Domain:

f (x) : [−1, 1] , f ′(x) : (−1, 1)

102
 y
6

2
x - sinHxL
1 - cosHxL
x
-6 -4 -2 2 4 6 sinHxL

-2

-4

-6

103
 y
6

2
x - sinH2 xL
1 - 2 cosH2 xL
x
-6 -4 -2 2 4 6 4 sinH2 xL

-2

-4

-6

104
 y
0.5
0.4
0.3
x2 - x3
0.2
H2 - 3 xL x
0.1
2-6x
x
-0.5 0.5 1.0
-0.1
-0.2

105
 y

0.10

0.05 -x3 + x4
x2 H-3 + 4 xL
x
-0.5 0.5 1.0 6 x H-1 + 2 xL
-0.05

-0.10

106
 y
5

3
1-x+x2
-1+x
2 1-4 x+2 x2
H-1+xL2

2
1
H-1+xL3

x
x
-2 -1 1 2 3

-1

-2

107
l’Hopital Rule

If f (c) = 0 AND g(c) = 0 with g ′(c) 6= 0, then:

f f ′(c)
lim = ′ (52)
x→c g g (c)

Demonstration (not proof):

f (c + dx) ≃ 0 + f ′(c) · dx

g(c + dx) ≃ 0 + g ′(c) · dx

′ ′ ′′  ′′
If f (c) = g (c) = 0, repeat to get f (c) g (c), etc.

108
Examples ”0/0”:
x2 − 1 2x 

lim = =2
x→1 x − 1 1 x=1
sin x cos x 

lim = =1
x→0 x 1 x=0
2

tan x 1/ cos x 
lim =  =1
x→0 x 1 
x=0
√ √ 
1+x−1 1/(2 1 + x)  1
lim =  =
x→0 x 1 
x=0 2

Caution: f (c) = g(c) = 0 is crucial! E.g.

sin x − 1 cos x − 0 

lim = −∞ , not =1
x→0 x 1 
x=0
109
 x − sin x 1 − cos x  sin x  1
 
lim = = =
x→0 x3 3x2 x→0 6x x→0 6

2 3
 
tan x − x 1/ cos x − 1  2 sin x/ cos x  1
lim 3
= 2
 =  =
x→0 x 3x 
x→0 6x 
x→0 3

√ √ 
1 + x − 1 − x/2 1/(2 1 + x) − 1/2 
lim 2
=  =
x→0 x 2x 
x→0
√ −3

−1/4 · ( 1 + x)  1
=  =−
2 
x→0 8

110
l’Hopital Rule for ”∞/∞”:

If f (x) → ∞ AND g(x) → ∞ as x → c with g ′(x) →

const 6= 0, then:

′

f f (x) 
lim = ′ (53)
x→c g

g (x) x→c

111
 n ln x 1/x
lim x ln x = lim −n = lim −n−1
=
x→0 + x→0 + x x→0 + −nx
1 n

=− x  = 0, n > 0
n x→0 +

lim x ln x = 0 (54)
x→0+

−n x ex ex
lim x e = lim n = lim = ...
x→∞ x→∞ x x→∞ nxn−1

ex
= lim →∞
x→∞ n!

112
”∞ · 0 → 0/0”:

sin(1/x) −(1/x2) cos(1/x)

lim x sin(1/x) = lim = lim 2
=1
x→∞ x→∞ 1/x x→∞ −1/x

”∞ − ∞ → 0/0”:
1 1  x − sin x  1 − cos x 
  
− = = =
sin x x x→0 x sin x x→0 x cos x + sin x x→0
sin x


=  =0
−x sin x + cos x + cos x x→0

113
”1∞”:
lim (1 + zx)1/x = ez
x→0+

”00”:
lim xx = 1
x→0+
(both via logarithm)

114
Applied Optimization
”Best rectangle”: P = 2(x + y) -fixed; S = xy = max.

S = x(P/2 − x) , dS/dx = P/2 − 2x = 0

x = P/4 = y

”Best can”: V = πr 2h - fixed; S = 2πrh+2πr 2 = min.

h = V /πr 2
V 1 V
 
S = 2π + r2 , dS/dx = − 2 + 2r = 0
πr 2π πr

V 1/3
 
r= , h = 2r
2π

115
 q
Rectangle in a semicircle: S = 2x R2 − x2 = max.
√
Ans.: x = R/ 2.

Refraction and reflection from Fermat principle (in

class).

Distance from a parabola (in class).

116
Using linearization for approximate calculations:
Let F (c) be easy to evaluate, then for small dx

F (c + dx) ≃ F (c) + F ′(c) · dx

1 x−2/3. Find 291/3 .

E.g. F (x) = x1/3 , F ′(x) = 3
Solution: Use c = 27:
1 2 83
(27 + 2)1/3 ≃ 3 + 27−2/3 · 2 = 3 =
3 27 27

117
Alternatively (1st step of Newtons method): f (x) =
x3 − 29. look for a root in linearized approximation;
starting guess x0 = 3

0 = f (x) ≃ f (x0) + (x − x0) · f ′(x0)

f (x )
x = x0 − ′ 0 = 3 − (−2)/(3 · 9) = 83/27
f (x0)
same thing.

118
 y

-2 + x2
2

x
0.5 1.0 1.5 2.0 2.5 3.0

-2

119
 y

-29 + x3
x
2.8 3.0 3.2 3.4

-2

-4

120
General Newton’s method

f (xn)
xn+1 = xn − ′ (55)
f (xn)

E.g., f = x2 − 2, starting guess x0 = 3/2

xn 1
xn+1 = + ,
2 xn
3 17 577 665857
 
, , ,
2 12 408 470832
”Errors”:

{0.25, 0.007, 10−6 , 10−12}

121
f = x3 − 29, starting guess x0 = 3:
29 2xn
xn+1 = +
3x2
n 3

83 1714381 15116218032546583823
 
3, , ,
27 558009 4920136460591269347
”Errors”:

(−2. , 0.05 , 10−5 , 10−12)

122
Approximating ln 3 = 1.09861 . . . with Newton

f (x) = ex − 3

f (xn)
xn+1 = xn − ′ = xn + 3e−xn − 1
f (xn)
If start with x0 = 1,
3 3
x1 = 1 + − 1 = ≃ 1.10364
e e
with error 5 · 10−3. Next, x2 ≃ 1.09862 with error
10−5

123
Antiderivatives and indefinite integrals:
F (x) is A.D. of f (x) if

(F )′ = f
Note: F (x) + C - also an A.D. All together - indefinite
R
integral, f (x) dx

E.g.
Z
dx = x + C (56)
1 2
Z
x dx = x + C (57)
2
dx 1
Z
=− +C (58)
x2 x

124
Tables (remember constant!):

1
Z
xn dx = xn+1 , n 6= −1 (59)
n + 1Z
dx/x = ln |x| (60)
Z
sin x dx = − cos x (61)
Z
cos x dx = sin x (62)
Z
dx/ cos2 x = tan x (63)
Z q
dx/ 1 − x2 = arcsin x (64)
Z
dx/(x2 + 1) = arctan x (65)
Z
ex dx = ex (66)

125
Expanding the Tables:

•
Z Z Z
(af (x) + bg(x)) dx = a f dx + b g dx + C

R
• If F (x) = f (x) dx , then
1
Z
f (kx) dx = F (kx) + C
k

E.g.
1 1
Z Z
sin(ωt) dt = − cos(ωt)+C , e−2x dx = − e−2x+C , . . .
ω 2
126
Finding C.
Let F (x) - some A.D. of f (x). We want to construct a
specific A.D. which satisfies F (x0) = A . The general
form is F (x) + C AND

F (x0) + C = A , i.e. C = A − F (x0)

Thus, we solved a differential equation (F )′ = f with
additional ”initial” (or ”boundary”) condition F (x0) =
A.

127
E.g., dx/dt = 4t3 + 4t , x(0) = 7
Z
x(t) = (4t3 + 4t) dt = t4 + 2t2 + C , C = 7

E.g., free fall with dv/dt = −g = const , v(0) = V0 ,

dy/dt = v , y(0) = Y0 . (2 equations)
Z
v(t) = (−g) dt = −gt + C , C = V0

1
Z
y(t) = v(t) dt = − gt2 + V0t + C1 , C1 = Y0
2

E.g., 1-d motion with a = 4e2t , v(0) = 7 , x(0) = 3.

v(t) = 2e2t + C , C = 7 − 2e0 = 5

x(t) = e2t + 5t + C1 , C1 = 3 − e0 − 5 · 0 = 2
128
Small x approximations (from l’Hopital)

(1 + x)k − 1 ≃ kx (67)
ex − 1 ≃ x , ax − 1 ≃ x ln a (68)
ln(1 + x) ≃ x (69)
sin x ≃ tan x ≃ x (70)
x − sin x ≃ x3/6 (71)
tan x − x ≃ x3/3 (72)
1 − cos x ≃ x2/2 (73)

129
More examples on limits
q
4x2 − 1
lim = −2
x→−∞ x−π

2sin x − 1
lim = ln 2
x→0 x

lim (1 + 2x)5/x = e10

x→0
πx
lim cos = −1
x→0 sin x
2/x
2x x = e6

lim e +e −1
x→0

130
 y
2

x
-2 -1 1 2

-1

-2

133
 y
4

x
-2 -1 0 1 2

134
 v v
1.0 1.0

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

t t
0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0

135
 y
4

x
0.5 1.0 1.5 2.0

136
Σ-notations
N
X
ak = a1 + a2 + . . . + aN
k=1
k - ”dummy” variable:
N
X N
X NX
−1
ak = ai = am+1
k=1 i=1 m=0

N
X
f (ak ) = f (a1) + f (a2) + . . . f (aN )
k=1

137
 N
X N (N + 1)
k = 1 + 2 + ... + N = (74)
k=1
2

Indeed, consider and add

1 + 2 + . . . + (N − 1) + N

N + (N − 1) + . . . + 2 + 1
Also,
N
N (N + 1)(2N + 1)
k2 = 12 + 22 + . . . + N 2 =
X

k=1
6

138
Area under a straight line
Consider v(t) = t (acceleration = 1 and v(0) = 0) and
look for displacement from t = 0 to arbitrary t (in the
picture, t = 1). Use discrete
t
tk = k∆t = vk , ∆t =
N
NX
−1
N (N − 1) 1 1
vk ∆t = (∆t)2 = t2(1 − )
k=1
2 2 N

139
Area under a parabola
Consider y = x2 and look for area between a = 0 and
arbitrary b > 0. Use discrete
2 b−a
xk = k∆x , yk = (k∆x) , ∆x =
N
NX
−1
3 N (N − 1)(2N − 1) 1 3 1 1
yk ∆x = (∆x) = b (1− )(1− )
k=1
6 3 N 2N

140
Riemann Sums
Partition

P = a = x0 < x1 < x2 < . . . < xN −1 < xN = b

∆xk = xk − xk−1 , ||P || = max [∆xk ]

xk−1 ≤ ck ≤ xk

N
X
R.S. : f (ck )∆xk (75)
k=1

f (ck ) = max f (x) , xk−1 ≤ x ≤ xk - upper sum.

f (ck ) = min f (x) , xk−1 ≤ x ≤ xk - lower sum.

141
 y
4

x
0.5 1.0 1.5 2.0

142
 y
1.0

0.5

x
1 2 3 4 5 6

-0.5

-1.0

143
Definite integral:

Z b N
X
f (x) dx = lim f (ck )∆xk = (76)
a ||P ||→0 k=1
N
X b−a
= lim f (ck )∆x , ∆x = (77)
N →∞
k=1
N

Z b
area = f (x) dx , f (x) ≥ 0 , b > a (78)
a

144
Mean Value Theorem:
If f (x) is continuous on [a, b] then at some a ≤ c ≤ b
1 b
Z
f (c) = f¯ = f dx
b−a a
Proof: From min-max inequality min f ≤ f¯ ≤ max f .
Then use I.V.T.

Rb
E.g., if a f dx = 0, there is a root of f

145
Fundamental Theorem of Calculus. Pt.I
If f (x) is continuous on [a, b] then F (x) = ax f (t) dt is
R

continuous and diff. on (a, b) with

d x
Z
f (t) dt = f (x) (79)
dx a

Part II (”Evaluation Theorem”):

Z b
f (x) dx = F (b) − F (a) = F (x)|ba (80)
a

where F is any A.D. of f .

146
E.g. (Pt. I):
d x
Z
t sin t dt = x sin x
dx a
(a does not matter!)
d a
Z
t sin t dt = −x sin x
dx x

d x2 dx2
Z
2 2
t sin t dt = (x sin x ) = ...
dx a dx

147
E.g. (Pt.II):
Z π
sin x dx = − cos x|π
0 =2
0
Z π/4
dx π/4
= tan x|0 =1
0 cos2 x
Z 1
dx 1 = π/4
= arctan x|0
0 x2 + 1
Z 2
dx/x = ln x|2
1 = ln 2
1

eαb − 1
Z b
αx 1 αx b
e dx = e |0 =
0 α α

148
The ”net change theorem”
Z b
F (b) − F (a) = F ′(x) dx
a

Z t
2
s(t2) − s(t1) = v(t) dt
t1

149
 Z b
area = |f (x)| dx
a
not convenient since don’t know A.D. for |f |.

Z 2π
2π
| sin x| dx = − cos x|π
0 + cos x|π = 4
0

f = x(x + 1)(x − 2) , F (x) = x4/4 − x3/3 − x2

Z 0 Z 2
area = f dx− f dx = F (0)−F (−1)−[F (2)−F (0)] = . . .
−1 0

150
 y

x
-8 -6 -4 -2 2

-2
-3 x - x2
0

-4

-6

-8

F (x) = −x3/3 − 3x2/2 ,

A = −[F (−3) − F (−8)] + F (0) − F (−3) − [F (2) − F (0)]

151
 y
0.6

0.4

0.2

 x¤ sgnHxL
5
-x +
x
-1.0 -0.5 0.5 1.0 1.5 2.0 0

-0.2

-0.4

-0.6

5 6/5
F (x) = x − x2/2
6
area = 2[F (1) − F (0)] − [F (2) − F (1)] = . . .

152
Substitution method
Z Z
f (u)u′(x) dx = f (u) du (81)
1
dx = d(kx + b) (82)
k
x dx = d(x2/2) (83)
dx/x2 = −d(1/x) (84)
dx/x = d(ln |x|) (85)
sin x dx = −d(cos x) (86)
cos x dx = d(sin x) (87)
dx/ cos2 x = d(tan x) (88)
1
eax dx = d(eax) (89)
a

153
 1 1
Z Z
f (kx+b) dx = f (kx+b) d(kx+b) = F (kx+b)+C
k k
12
Z p
kx + b dx = (kx + b)3/2 + C
k3
1
Z
sec2(kx + b) dx = tan(kx + b) + C
k

1 x2
Z Z
xe x2 dx = x2 2
e d(x /2) = e + C
2
1 x3
Z Z
x2 e x3 dx = x3 3
e d(x /3) = e + C
3

154
 cos x dx du
Z Z Z
I= sec x dx = 2
= 2
, u = sin x
cos x 1−u
1 1 1 1  1 + u 
Z    
I= + du = ln  +C
2 1−u 1+u 2 1−u

u − 1√ u
Z p Z
I= x 2x + 1 dx = u d , u = 2x+1 , x = (u−1)/2
2 2
1 2 5/2 2 3/2
I= ( u − u ) + C = ...
4 5 3
u = cos x:
sin x dx du
Z Z Z
tan x dx = =− = − ln |u|+C = ln | sec x|+C
cos x u
155
Substitution method for definite integrals
Z b Z u(b)
f (u)u′(x) dx = f (u) du (90)
a u(a)

(x3 + 1) 1
Z 1 Z 1 q
1 2 3 2 3/2
q 
x2 x3 + 1 dx = 3
x + 1d = · (x + 1) 3/2  = 2
−1 −1 3 3 3 
−1 9
or u = x3 + 1 , u(−1) = 0 , u(1) = 2
1 2√ 1 2 3/21 2 3/2
Z 
u du = u = 2
3 0 33 
0 9

156
 y
f HuL
1.0
1.0

0.8 0.8

0.6 0.6
2 3 u
x 1+x
3
0 0
0.4 0.4

0.2 0.2

x u
-1.0 -0.5 0.5 1.0 0.5 1.0 1.5 2.0

157
u = cot x:
Z π/2 Z cot π
2 1 1
cot x csc2 x dx = − u du = − u2|0
1 =
π/4 cot π4 2 2

u = cos x:
Z π/4 Z π/4 Z cos(π/4) Z 1/√2
sin x dx du
tan x dx = =− =− √ ... = 0
−π/4 −π/4 cos x cos(−π/4) u 1/ 2

158
 y
y
1.0
4

3 0.5

5 x2 1
2+ - x4 sinHxL + 3
sinH3 xL
2
2 x
0 -3 -2 -1 1 2 3 0

1
-0.5

x
-1.5 -1.0 -0.5 0.5 1.0 1.5 -1.0

Z a Z a
f (x) dx = 2 f dx , (f − even)
−a 0
Z a
f (x) dx = 0 , (f − odd)
−a

159
Proof
Z a Z 0 Z a
I= f (x) dx = f (x) dx + f (x) dx = I1 + I2
−a −a 0
Z 0 Z a Z a
I1 = f (x) dx = f (−u) du = ± f (u) du =
−a 0 0

= ±I2 , (” + ” f or even , ” − ” f or odd)

E.g.
Z 1 Z 1
(x4−3x3+4x2+x−2) dx = 2 (x4+4x2−2) dx = . . .
−1 0

160
Area between two curves
Z b
A= (f − g) dx , f (x) ≥ g(x)
a
y
3

2 - x2
x -x
-1.0 -0.5 0.5 1.0 1.5 2.0

-1

-2

-3

161
 f = 2 − x2 , g = −x
Intersection:

f (x) = g(x) , x = −1 and x = 2

Z 2 2
A= 2 3 2
(2 − x + x) dx = (2x − x /3 + x /2)

=
−1 −1

1 3 1 2
= 2(2 + 1) − (2 + 1) + (2 − 1) = . . . > 0
3 2

162