Professional Documents
Culture Documents
Lecture Notes 230
Lecture Notes 230
David M. McClendon
Department of Mathematics
Ferris State University
2016 edition
2016
c David M. McClendon
1
Contents
Contents 2
1 Review of Calculus I 5
1.1 Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3 The definite integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4 Homework exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4 Improper integrals 61
4.1 Boundedness vs. unboundedness . . . . . . . . . . . . . . . . . . . . . 61
4.2 Horizontally unbounded regions . . . . . . . . . . . . . . . . . . . . . 63
4.3 Vertically unbounded regions . . . . . . . . . . . . . . . . . . . . . . . 66
4.4 Theoretical approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.5 Homework exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
2
Contents
3
Contents
Index 263
4
Chapter 1
Review of Calculus I
2. What are some typical kinds of problems you learn how to solve / mathe-
matical procedures you learn in Calculus I?
5
Most of Calculus I theory in one chart
CALCULUS
OBJECT
MOTIVATING
PROBLEM
APPROXIMATION
OF THE
SOLUTION
HOW THE
APPROXIMATION
IMPROVES
THEORETICAL
SOLUTION
HOW THE
THEORETICAL
SOLUTION
IS COMPUTED
IN PRACTICE
APPLICATIONS
6
1.1. Limits
What is the difference between a math problem that is a calculus problem and a
math problem that is NOT a calculus problem?
A calculus problem is one that contains a limit; a non-calculus problem con-
tains no limit.
What is calculus?
Calculus is the study of limits.
DERIVATIVE - APPLICATIONS
*
LIMIT - INTEGRAL - APPLICATIONS
HH
HH
H
j
H
? - ?
1.1 Limits
Recall: To say
lim f (x) = L
x→a
means that as x gets closer and closer to a, then f (x) gets closer and closer to L.
This suggests that the graph of f looks like one of the following three pictures:
The graph on the left is “continuous” at a; the other two graphs are not. More
precisely,
7
1.1. Limits
Theorem 1.2 Any function which is the quotient of functions made up of powers of
x, sines, cosines, arcsines, arctangents, exponentials and/or logarithms is continuous
everywhere except where the denominator is zero.
This theorem suggests that to evaluate most limits, you should start by plug-
ging in a for x. If you get a number, that is usually the answer.
2. lim x+3
x−2
x→4
2 −3x−10
4. lim x x−5
x→5
When you plug in a for x and you get 0 in a denominator, you have to work a
bit harder to evaluate a limit.
nonzero
• When you get 0
:
x−5
Example 3 above: lim (x−2) 2
x→2
8
1.1. Limits
2 −3x−10
Example 4 above: lim x x−5
x→5
Theorem 1.3 (L’Hôpital’s Rule) Suppose f and g are differentiable functions. Sup-
pose also that either
Then:
f (x) L f 0 (x)
lim =lim .
x→a g(x) x→a g 0 (x)
Limits at infinity
We can also take a limit of a function as x → ∞. To say lim f (x) = L means that
x→∞
as x grows larger and larger without bound, then f (x) approaches L. Graphically,
this means f has a horizontal asymptote (HA) y = L:
9
1.1. Limits
√
(
∞ c ∞ if c > 0
∞=∞ ln ∞ = ∞ e =∞ ∞ =
0 if c < 0
c ∞ c
=0 = ±∞ = ±∞ so long as c 6= 0
∞ 0 0
In the last two situations, you need to analyze the sign carefully to determine
whether the answer is ∞ or −∞.
Warning: The following expressions are indeterminate forms, i.e. they can
work out to be different things depending on the context (these are evaluated using
L’Hôpital’s Rule):
0 ∞
0·∞ ∞−∞ 1∞ ∞0 00
0 ∞
lim lnxx .
Example 5: Evaluate the limit x→∞
3
lim 4x
Example 6: Evaluate the limit x→∞ +2x+1
2x3 +x2 +2
.
The last example generalizes into the following principle which has many use-
ful applications:
10
1.2. Derivatives
Then:
1.2 Derivatives
Definition 1.5 (Limit definition of the derivative) Let f : R → R be a function
and let x be in the domain of f . If the limit
f (x + h) − f (x)
lim
h→0 h
exists and is finite, say that f is differentiable at x. In this case, we call the value of
df dy
this limit the derivative of f and denote it by f 0 (x) or dx or dx .
Differentiable functions are smooth, i.e. are continuous and do not have sharp
corners, vertical tangencies or cusps.
11
1.2. Derivatives
• text
–
–
–
–
12
1.2. Derivatives
4 dy
Example: Suppose y = x2
− 2 ln x + 9e2x . Find dx
.
Example: Find the slope of the line tangent to the graph of f (x) = 3 cos x+2 sin x
at x = π4 .
Derivatives have many applications. The most important (for Math 230 pur-
poses) is that given a differentiable function f , you can approximate values of f
near a using the tangent line to f at a:
13
1.3. The definite integral
For values of x near a, f (x) ≈ L(x); approximating f (x) via this procedure is called
linear approximation.
You can also conclude many things about the graph of f from looking at its
derivative and its higher-order derivatives, as defined below:
• The zeroth derivative of f , sometimes denoted f (0) , is just the function f itself.
dy
• The first derivative of f , sometimes denoted f (1) or dx
, is just f 0 .
d2 y
• The second derivative of f , denoted f 00 or f (2) or dx2
, is the derivative of f 0 :
f 00 = (f 0 )0 .
dn y
• More generally, the nth derivative of f , denoted f (n) or dxn
,is the derivative of
f (n−1) :
f (n) = ((((f 0 )0 ) · · ·0 )0
The first derivative of a function measures its tone (i.e. whether it is increasing
or decreasing). The second derivative of a function measures its concavity (i.e.
whether its graph is cupped upward or downward). In Math 230, we will see
what the higher-order derivatives of f have to do with the function.
14
1.3. The definite integral
Note: In Math 230, the limit above always exists (but it doesn’t always exist for
crazy functions f ... take Math 430 for more on that).
Every continuous function has an antiderivative (although you may not be able
to write its formula down); any two antiderivatives of the same function must dif-
fer by a constant (so if you know one antiderivative, you know them all by adding
a +C to the one you know).
Despite the similar notation, f (x) dx and ab f (x) dx are very different objects.
R R
15
1.3. The definite integral
Exercise: Without looking anything up, evaluate each of the following indefi-
nite integrals, if they are “easy” to solve. If the integral isn’t easily solved, put a
“?”. The idea here is that an integral which is “easy” should use facts from alge-
bra regarding exponents, logarithms, etc. and the rules memorized in Calculus I.
“Easy” integrals do not use sophisticated calculus techniques or trig identities.
R R
1. 0 dx = 22. cot x dx =
R
2. 1 dx = sin2 x dx =
R
23.
R
3. 5 dx = 24.
R
cos2 x dx =
x4 dx =
R
4. 25.
R
sec2 x dx =
R
5. x dx = 26.
R
csc2 x dx =
R 1
6. dx =
tan2 x dx =
R
x 27.
R 1
7. x3
dx =
cot2 x dx =
R
28.
x−5 dx =
R
8. R
29. sec x tan x dx =
R 1
9. x2 +1
dx = R
30. sec x cot x dx =
R 1
10. x3 +1
dx = R
R √ 31. csc x tan x dx =
11. x dx = R
R √ 32. csc x cot x dx =
12. 3 x dx =
R √
R
3
33. csc x tan x dx =
13. x2 dx =
R
R 2/5
34. csc x sec x dx =
14. x dx =
R x
R −1/2 35. e dx =
15. x dx =
R x2
R 1 36. e dx =
16. √ 5 x
dx =
R x
17.
R
sin x dx = 37. 2 dx =
R
18.
R
cos x dx = 38. ln x dx =
R
19.
R
tan x dx = 39. log10 x dx =
R
40. arctan x dx =
R
20. sec x dx =
R R
21. csc x dx = 41. arcsin x dx =
16
1.3. The definite integral
We know a little about how to integrate functions whose pieces are integrals
we memorize, but we don’t know as much about how to compute integrals as we
know about how to compute derivatives:
17
1.3. The definite integral
18
1.4. Homework exercises
1. 12.
x−1 e2x − 1
lim 2 lim
x→1 x + 1 x→0 x
2. 13.
lim 7 sin x
x→4 lim
x→0 x
3. 14.
x2 − 9
lim 2 lim ln x
x→3 x + x − 12 x→∞
4. 15.
x+2 lim ln x
lim+ 2 x→0+
x→2 x − 4
5. 16.
lim e−x lim ex
x→∞ x→0
6. 17.
1
x
+ 13 lim e1/x
lim x→0+
x→−3 1 + 1
x+2
18.
7. lim sin x
x x→∞
lim 2
x→∞
19.
8. lim arctan x
3x x→∞
lim
x→∞ x2 + 2
20.
9. lim e−1/x
−2 x→0+
lim
x→∞ x 21.
10. (
x2 x ≤ 1
2
4x − 3x + 1 lim f (x) where f (x) = .
lim x→1+ 5x x > 1
x→∞ 3 − 2x2
11. 22.
4x2 |x|
lim x lim
x→∞ e − 2 x→0 x
dy
23. Find dx
if y = x3 − 3x4 + 2.
24. Find f 0 (x) if f (x) = x1 .
19
1.4. Homework exercises
28. If the position of an object at time t is given by f (t) = (3 ln t)2 , find the velocity
and acceleration of the object at time t = e.
ex
29. If f (x) = 3 ln x + 5
, find f 00 (x).
x2 −1
30. If f (x) = x
, find f 0 (x).
√ 1 dy
31. Let y = 3 x − x3/7 + 3 2.
√
x
Find dx
.
sin x df
32. Let f (x) = 7ex − 4
. Find dx
.
d
33. Find dx
(2 tan x − 3 cos x).
34. Find the derivative of f (x) = 4 cot x + 2 sec x − 1.
35. Find f 0 (x) if f (x) = e4x .
36. Find h0 (x) if h(x) = x
tan x
.
√
37. Find the derivative of f (x) = sin x.
dy
38. Find dx
if y = 2x4 sin x.
39. Find the second derivative of f (x) = cos(3x2 ).
40. Find the ninth derivative of f (x) = 4 sin x.
x2 −3x+4
41. Find f 0 (x) if f (x) = 2x3 −x+1
.
42. Find the derivative of f (x) = x ln(3x2 + 1).
43. Find the derivative of f (x) = 14 arctan x2 .
44. Find the zeroth derivative of f (x) = 8x7 + 2 cot 4x.
45. Find the slope of the line tangent to y = 2x3 − 3x when x = 2.
46. Find the equation of the line tangent to y = 4x(x2 − 3)4 when x = 2.
dy
47. Find dx
if y = xex .
48. Find f 0 (0) if f (x) = |x|.
49. Let y = 4x2 . Find dy.
20
1.4. Homework exercises
6 sec2 x dx
R 5 R
54. x
dx 59.
R −6
(4ex + 1) dx
R
55. 60. x2 +1
dx
R
61. Compute (sec x tan x + csc x cot x) dx.
R √
62. Compute 03 9 − x2 dx.
Hint: Think about the shape of the graph of the integrand.
R4
63. Compute −3 |x| dx.
Hint: Think about the shape of the graph of the integrand.
64. Find the derivative of f (x) = x2 sin x. What does your answer tell you about
the value of Z π
2x sin x + x2 cos x dx?
0
Rx
65. Find the derivative of the function F (x) = 2 3t2 dt.
Hint: Use the part of the Fundamental Theorem of Calculus not mentioned
in these notes.
66. Find the area under the graph of the function f (x) = √8 from x = 1 to x = 9.
x
67. Suppose that an object’s velocity at time t is given by v(t) = 4 sin t. Find the
displacement of the object between times 0 and 2π 3
.
68. Suppose that the acceleration of an object is given by a(t) = 48t. If the velocity
of the object at time 0 is 12 and the position of the object at time 0 is 10, what
is the position of the object at time 3?
21
1.4. Homework exercises
Answers
1. 0 25. 5
2. 7 26. 3
−1
3. 6 27. 12
7
18
28. v(e) = ; a(e) = 0
4. ∞ e
−3 ex
29. x2
+ 5
5. 0
x2 +1
1
30. x2
6. 9
31. 3
√
2 x
− 37 x−4/7 − 23 x−5/3
7. ∞ cos x
32. 7ex − 4
8. 0
33. 2 sec2 x + 3 sin x
9. 0 34. −4 csc2 x + 2 sec x tan x
10. −2 35. 4e4x
tan x−x sec2 x
11. 0 36. tan2 x
37. √1 cos x
12. 2 2 sin x
22
1.4. Homework exercises
1
51. 2
61. sec x − csc x + C
52. x3 − 41 x8 + 5x + C 62. 9π
4
53. 2 3/2
9
x − 35 x−3 + C 63. 25
2
54. 5 ln x + C
64. f 0 (x) = 2x sin x + x2 cos x; it tells
55. 4ex + x + C you that the value of the integral
π
1 2
is [x2 sin x]0 = 0.
56. 2
x − 2x + ln x + C
65. 3x2
57. C
58. 3 66. 32
23
Chapter 2
In the next two chapters we will discuss various methods to compute antideriva-
tives/integrals. The general idea is that there are a variety of techniques which
work in various circumstances; one needs to learn which technique to use in which
situation.
24
2.1. Integrals to memorize
d 2 d
(x ) = 2x and (sin x) = cos x.
dx dx
Every derivative rule you learn is also an integration rule; for example the above
two rules translate as
and
We also learn the following rules, called the linearity rules for integration, which
allow us to split integrals into pieces:
Z Z Z
[f (x) + g(x)] dx = f (x) dx + g(x) dx;
Z Z Z
[f (x) − g(x)] dx = f (x) dx − g(x) dx;
Z Z
[k · f (x)] dx = k f (x) dx for any constant k.
25
2.1. Integrals to memorize
26
2.1. Integrals to memorize
27
2.2. Rewriting the integrand
28
2.3. Elementary u-substitutions
by setting u = g(x).
by setting u = g(x).
Notice that for this formula to work, the integrand must have the form
f 0 (g(x)) · g 0 (x).
That means that for an elementary u−substitution to be a valid integration tech-
nique, the integrand must consist of two terms multiplied together, where one
29
2.3. Elementary u-substitutions
30
2.3. Elementary u-substitutions
31
2.3. Elementary u-substitutions
R
Example 12: We know that cos x dx = sin x + C. Evaluate the following inte-
grals: Z
cos 2x dx
−5
Z
cos x dx
3
Z
cos (7x − 4) dx
Z
cos (3 − x) dx
32
2.3. Elementary u-substitutions
33
2.3. Elementary u-substitutions
34
2.4. More complicated u-substitutions
35
2.4. More complicated u-substitutions
36
2.4. More complicated u-substitutions
37
2.4. More complicated u-substitutions
Sometimes you can use trig identities to rewrite an integrand before using a
complicated u−substitution:
38
2.4. More complicated u-substitutions
Here are two last examples which illustrates how if you are just willing to try
something, sometimes things work out:
39
2.4. More complicated u-substitutions
40
2.5. Homework exercises
1. 12. 24.
Z
2
Z
sin x Z 1 √
t2 t − dt √ dx x x + 1 dx
t cos x
0
2. 13. Z
Z 4 √ e7−3x dx 25.
(3 − x) x dx Z
0
14. x2 (x + 3)8 dx
3. Z
1
Z dx
(1 + 2x ) dx 2 2 7x + 1 26.
Z
15. Z cos2 3x dx
4. sin 4x dx
2
1
Z
x 1+ dx 16. 27.
x
2
Z Z
5. cos x dx sin2 4x dx
Z π/3 3
3 cot2 x dx 17.
π/4 Z 28. Z
2 sec2 3x dx sin5 x dx
6.
Z
(x − 7)2 18. 29.
dx
Z
x/2
x e dx Z
cos3 x sin x dx
7. 19.
Z √ 2
Z
x x3 − 1 dx csc πx cot πx dx 30.
Z
8. 20. sin4 2x cos 2x dx
Z
x2 + 1 Z
3x + 2
dx dx
x3 + 3x − 2 5x − 1 31.
9. 21. Z
1 + sin x
2x
Z
1 dx
Z
e dx cos x
dx (4x + 5)3
e2x + 1
32.
10. 22. Z
1
Z 3 2x
e +1
Z 3
x2 dz
dx dx cos z − 1
0 e2x 2 x−1
11. 23. 33.
2x ln x2
Z Z Z
cot x dx dx dx
x−4 x
41
2.5. Homework exercises
R
34. a) Evaluate the integral sin x cos x dx by performing the u−substitution
u = sin x.
b) Evaluate the integral of part (a) by performing the u−substitution u =
cos x.
c) Reconcile the (apparently) different answers you get to parts (a) and (b)
of this question.
Answers
Note: with indefinite integrals, answers can sometimes vary a little bit.
1 4 3 2
1. 4
t − t2 + C 16. 2
sin 3
x +C
16
2. 5 17. 2
tan 3x + C
3
4 5
3. 5
x + 43 x3 + x + C 18. 2ex/2 + C
1 2
4. 2
+ 2x + ln x + C
x 19. −1
csc πx + C
π
√
5. 3 − 3 − π4 20. 3
5
x + 13
25
ln(5x − 1) + C
1 2 −1
6. 2
x − 14x + 49 ln x + C 21. 8
(4x + 5)−2 + C
2
7. 9
(x3 − 1)3/2 + C 22. 7
2
+ ln 2
1
8. 3
ln(x3 + 3x − 2) + C 23. 2x + 8 ln(x − 4) + C
1
√
9. 2
ln(e2x + 1) + C 4
24. 15 4
+ 15 2
10. 7
2
− 12 e−6 25. 1
11
(x + 3)11 − 53 (x + 3)10 + (x + 3)9 + C
1 1
11. ln(sin x) + C 26. 2
x + 12
sin 6x + C
√ 1 1
12. −2 cos x + C 27. 2
x − 16
sin 8x + C
−1 7−3x −1
13. 3
e +C 28. 5
cos5 x + 32 cos3 x − cos x + C
1 −1
14. 7
ln(7x + 1) + C 29. 4
cos4 x + C
−1 1
15. 4
cos 4x + C 30. 10
sin5 2x + C
42
2.5. Homework exercises
Hint: First, rewrite the integrand by multiplying through the numerator and
denominator by (cos z + 1).
33. ln2 x + C
Hint: First, rewrite ln x2 as 2 ln x.
43
Chapter 3
Intermediate integration
techniques
(f g)0 (x) =
44
3.1. Integration by parts
R
The IBP formula
R
is used (usually) Rto find integrals of the form u dv (you choose
u and dv) where v du is easier than u dv.
R
Example 1: x sin x dx
Thought Process:
Solution:
x2 ex dx
R
Example 2:
Solution:
45
3.1. Integration by parts
sine or cosine, you would (at least theoretically) use integration by parts n times.
R1 2 x
Example 3: x e dx
0
Solution:
R
Example 4: ln x dx
Solution:
R R
(The integrals arctan x dx and arcsin x dx are similar.)
46
3.1. Integration by parts
Specific guidelines:
MEDIUM PRIORITY u:
3x2 arctan x dx
R
(a)
R
(b) x ln x dx
R
(c) 2x sin 4x dx
R 3x
(d) e cos 2x dx
R ln x
(e) x
dx
R ln x
(f) x2
dx
4x3 e2x dx
R
(g)
R x2 +3x−3
(h) 2x+3
dx
R 2x+3
(i) x2 +3x−3
dx
47
3.2. Partial fractions
3x−5 2 1
Goal: Rewrite expressions like x2 −4x+3
as x−3
+ x−1
.
Examples:
• x4 − 3x + 1 is a degree 4 polynomial
• x2 − 3x9 − 7x + 5x6 is a degree 9 polynomial
7 √
• x x+2 , x, x1 , sin x, 4xπ , ... are not polynomials
• x2 + A is an irreducible polynomial if A > 0
• x2 − 4 is not irreducible
• Ax ± B is irreducible
In the same way that every whole number factors into a product of primes, we
have:
48
3.2. Partial fractions
STEP 3: Write the equation from step 2 so that it has common denominator
49
3.2. Partial fractions
50
3.2. Partial fractions
51
3.2. Partial fractions
2x3 − x2 − x + 1
x2 (x − 1)2
52
3.2. Partial fractions
It is hard to write down a “one-size fits all” statement; here are some examples:
p(x)
q(x)
GUESSED FORM OF DECOMPOSITION
6x−1
(x−5)(x+2)
−4
x(x−3)(x+4)
2
(x−4)2 (x+6)
7
x4 (x−1)3 (x+2)
x−1
(x2 +5)(x−2)
x3 −3x+4
(x2 +1)3 x2 (x−7)(x2 +6)
General guidelines:
1. Make sure the denominator is factored completely first.
2. For every linear term (x ± a)r in the denominator, the guessed form needs r
terms of the form
A B C const
+ + + ... +
x ± a (x ± a)2 (x ± a)3 (x ± a)r
3. For every unfactorable quadratic (x2 +ax+b)r in the denominator, the guessed
form needs r terms of the form
Ax + B Cx + D Ex + F const x + const
+ 2 + 2 + ... +
x2 + ax + b (x + ax + b)2 (x + ax + b) 3 (x2 + ax + b)r
53
3.2. Partial fractions
Example: Z
x
dx
(x − 2)3
Solution # 1: Use partial fractions:
x A B C
= + +
(x − 2)3 x − 2 (x − 2) 2 (x − 2)3
x A(x − 2)2 B(x − 2) C
= + +
(x − 2)3 (x − 2)3 (x − 2)3 (x − 2)3
x = A(x − 2)2 + B(x − 2) + C
x = A(x2 − 4x + 4) + Bx − 2B + C
x = Ax2 + (B − 4A)x + (4A − 2B + C)
Solution # 2:
54
3.2. Partial fractions
Z
1
dx = ln |x ± a| + C
x±a
Z
1 1
dx = ln |ax ± b| + C
ax ± b a
1 −1 1
Z
n
dx = + C (if n 6= 1)
(x − a) n − 1 (x − a)n−1
Z
1
dx = arctan x + C
x2 + 1
Z
1 1 x
2 2
dx = arctan + C
x +a a a
55
3.3. Summary of integration techniques
2. If the integrand contains terms which are added or subtracted, consider using
linearity rules to split the integrand into pieces you can “just write the answer
to”.
3. See if you can rewrite the integrand using algebra, log rules or a trig identity.
5. If the integrand contains terms which are multiplied together, where one
part is the derivative of something in the other part (up to a constant), use
a u−substitution.
6. If the integrand has terms multiplied together which are unrelated, try inte-
gration by parts.
7. If the integrand is a rational function where the denominator factors (and the
degree of the denominator is greater than the degree of the numerator), use
partial fractions to decompose the integrand.
56
3.5. Homework exercises
1. Z 6. Z
−3x
xe dx x cos x dx
2. 7. Z π/2
Z
4x
dx x sin 2x dx
ex 0
3. Z 8. Z 1
3 x4
x e dx 2xe−x/2 dx
0
4. Z 9. Z
x ln(x + 5) dx arctan x dx
5. Z 10. Z
−2
x ln x dx x arctan x dx
11. 16.
Z
x arctan x
√
Z
dx dx
2 + 3x x2 + 1
12. Z
x2 cos x dx 17. Z
1
dx
13. Z x ln x
x ln2 x dx
18. Z
14. Z 3x2 e2x−1 dx
arcsin x dx
15. Z 19. Z
sin x cos x dx ex cos(3x) dx
In Problems 20-23, write the “guessed” form of the partial fraction decomposi-
tion of the given expression. You do not need to solve for the constants.
1 1 A B
Example: Solution: = +
x2 − 1 x2 − 1 x−1 x+1
57
3.5. Homework exercises
4 3x
20. x2 −3x
22. x2 +4x−5
7x+2 2
21. x3 −5x2 −24x
23. 3x2 +9x−54
24. 26.
Z
1
Z
5−x
dx dx
x2 − 16 2x2 + x − 1
25. 27.
Z
3 Z
5x2 − 12x − 12
dx dx
x +x−2
2 x3 − 4x
In Problems 28-33, write the “guessed” form of the partial fraction decomposi-
tion of the given expression. You do not need to solve for the constants.
2x+3 1
28. x(x2 +1)
31. (x−2)2 (x2 +4)2
4 x+1
29. (x−3)(x+2)2 32. x3 (x2 +x+1)(x−7)
5−x2 7
30. x3 (x−1)2
33. (x+4)3
34. 36.
Z
4x2 + 2x − 1
Z
x2 − x + 9
dx dx
x3 + x2 (x2 + 9)2
35. 37.
Z
x2
Z 5
x−1
dx dx
x4 − 2x2 − 8 1 x2 (x + 1)
integral 2x−3
R
38. a) Evaluate the (x−1)3
dx by performing a u−substitution.
b) Evaluate the integral from part (a) via partial fractions.
c) Verify that the answers you get to parts (a) and (b) are equal.
d) Evaluate the integral using Mathematica. Which method do you think
Mathematica used to compute the integral?
39. In the “Elementary integral formulas used with partial fractions” subsection
on p. 54, we asserted that
Z
1 1 x
dx = arctan + C.
x2 +a 2 a a
Show that this formula is correct. Hint: Rewrite the integrand by factoring
out a2 from the denominator and then use the Linear Replacement Principle.
58
3.5. Homework exercises
Answers
−1
1. 3
xe−3x − 19 e−3x + C
2. −4xe−x − 4e−x + C
1 x4
3. 4
e +C
−1 3
4. 4
x + 52 x + 12 x2 ln(x + 5) − 25
2
ln(x + 5) + C
−1
5. x
− x1 ln x + C
6. x sin x + cos x + C
π
7. 4
8. 8 − 12e−1/2
9. x arctan x − 21 ln(x2 + 1)
1 2 −1
10. 2
x arctan x + 2
x + 12 arctan x + C
11. u-SUB; u = 2 + 3x
17. u-SUB; u = ln x
7x+2 A B C
21. x3 −5x2 −24x
= x
+ x−8
+ x+3
3x A B
22. x2 +4x−5
= x+5
+ x−1
2 A B
23. 3x2 +9x−54
= x+6
+ 3x−9
1
24. 8
ln(x − 4) − 81 ln(x + 4) + C
59
3.5. Homework exercises
A B C
29. x−3
+ x+2
+ (x+2)2
A B C D E
30. x
+ x2
+ x3
+ x−1
+ (x−1)2
A B Cx+D Ex+F
31. x−2
+ (x−2)2
+ x2 +4
+ (x2 +4)2
A B C Dx+E F
32. x
+ x2
+ x3
+ x2 +x+1
+ x−7
A B C
33. x+4
+ (x+4)2
+ (x+4)3
1
34. x
+ 3 ln x + ln(x + 1) +C
√
35. 6
2
arctan √x2 + 16 ln(x − 2) − 61 ln(x + 2) + C
36. 1
2
(x2 + 9)−1 + 13 arctan x3 + C
−4
37. 5
+ ln 25
9
60
Chapter 4
Improper integrals
Since f is continuous on [a, b], it must be that the graph of f encloses a bounded
region (in the sense that one can draw a box around it):
Rb
This guarantees that a f (x) dx < ∞.
61
4.1. Boundedness vs. unboundedness
Notice that in the previous picture, the region obtained was unbounded, but
in the direction it was unbounded, the width/thickness of the region decreases
to zero. Another situation where the width/thickness of an unbounded region
decreases to zero is when the region is described by a graph with an asymptote:
62
4.2. Horizontally unbounded regions
this limit exists and is equal to a finite number L, we write a∞ f (x) dx = L nd say
R
If
R∞
Ra∞
f (x) dx converges (to L). If this limit does not exist or is equal to ±∞, we say
a f (x) dx diverges.
1 b
63
4.2. Horizontally unbounded regions
1 b
64
4.2. Horizontally unbounded regions
The results of the previous problem are summarized in the following theorem,
which should be memorized:
65
4.3. Vertically unbounded regions
66
4.3. Vertically unbounded regions
this limit exists and is equal to a finite number L, we write ac f (x) dx = L and say
R
If
Rc
Rac
f (x) dx converges (to L). If this limit does not exist or is equal to ±∞, we say
a f (x) dx diverges.
a b c
The next situation is when the vertical asymptote is on the left-hand edge of the
region whose are you want to compute:
Definition 4.4 (Improper integrals for vertically unbounded regions II) (VA on
the left-hand edge of the region) Let f (x) beR continuous on (a, c] and suppose
lim+ f (x) = ±∞. Define the improper integral ac f (x) dx to be
x→a
Z c Z c
f (x) dx = lim+ f (x) dx.
a b→a b
this limit exists and is equal to a finite number L, we write ac f (x) dx = L and say
R
If
Rc
Rac
f (x) dx converges (to L). If this limit does not exist or is equal to ±∞, we say
a f (x) dx diverges.
67
4.3. Vertically unbounded regions
a b c
Definition 4.5 (Improper integrals for vertically unbounded regions III) (VA
in middle of region) Let f (x) be continuous on [a, c] except at a single point b ∈
(a, c). Suppose lim+ f (x) = lim− f (x) = ∞ or lim+ f (x) = lim− f (x) = −∞.
x→b
Rx→b x→b x→b
the improper integral ac f (x) dx is said to converge only if ab f (x) dx and
R
Then
Rc
b f (x) dx converge (in the sense of the previous definitions). In this case we set
Z c Z b Z c
f (x) dx = f (x) dx + f (x) dx
a a b
where the two integrals on the right side are defined as in the preceding definitions.
a b c
68
4.3. Vertically unbounded regions
69
4.3. Vertically unbounded regions
70
4.4. Theoretical approaches
The first “trick” is that when integating over a horizontally unbounded region,
the starting index of the integral doesn’t affect whether or not the integral con-
verges.
A picture to explain:
If f is positive and a < b, then (assuming these integrals converge) it is clear that
Z ∞ Z ∞
f (x) dx > f (x) dx
a b
71
4.4. Theoretical approaches
The next set of results describe how one can “split up” improper integrals
whose integrands are composed of terms added/subtracted together:
R∞
Theorem 4.7 (Linearity of improper integrals I) Suppose
R∞ a f (x) dx = L and
a g(x) dx = M . Then:
R∞
1. a [f (x) + g(x)] dx = L + M ;
R∞
2. a [f (x) − g(x)] dx = L − M ;
R∞
3. a kf (x) dx = kL for any constant k.
R∞
Theorem 4.8 (Linearity of improper integrals II) Suppose
R∞ a f (x) dx = L but
a g(x) dx diverges. Then:
R∞
1. a [f (x) + g(x)] dx diverges;
R∞
2. a [f (x) − g(x)] dx diverges;
R∞
3. a kg(x) dx diverges, for any constant k 6= 0.
The preceding three theorems can be summarized in the following three “morals”,
which we will see again and which hold throughout mathematics:
1.
2.
3.
72
4.4. Theoretical approaches
R∞ 1 3
(b) 6 5x2
− x3/2
dx
Our last theoretical result is based on a very important and very general math-
ematical idea of reasoning by way of inequalities:
0 ≤ f (x) ≤ g(x)
A picture to explain:
73
4.4. Theoretical approaches
74
4.4. Theoretical approaches
4+?
where , 4 and ? are all positive quantities. Then, one can start with one of
these two inequalities
≤ or ≤
4+? 4 4+? ?
(The reason these inequalities are valid is that removing positive terms from
the denominator makes the denominator smaller, which makes the entire
fraction bigger.)
75
4.4. Theoretical approaches
4−?
where , 4 and ? are all positive quantities. Then, one can start with this
inequality:
≥
4−? 4
and try to apply the Comparison Test.
(The reason this inequality is valid is that removing negative terms from the
denominator makes the denominator bigger, which makes the entire fraction
smaller.)
−1 ≤ cos ≤ 1 ( or − 1 ≤ sin ≤ 1)
76
4.5. Homework exercises
1. Z ∞ 4. Z ∞
−4 2
x dx dx
1 1 x2 +1
2. Z ∞ 5.
1 Z ∞
ln x
√ dx dx
8
3
x 1 x
3. 6. Z ∞
Z ∞
−x
2
xe dx dx
0 3 (x + 3)5/3
R∞ 1
7. In class we showed that 1 x dx diverges. Have Mathematica evaluate this
integral. Does Mathematica recognize that this integral diverges?
8. Use Mathematica to find the values of
Z ∞
xn e−x dx
0
Then, sketch a picture (similar to those on pages 66-67) which reflects how
the limit is being used to evaluate the improper integral.
R2 1
a) 1 x ln x dx
R1 x
b) 0 x2 −1 dx
R∞ 1
c) 0 xex dx
R5 1
d) 3 (x−5)2 dx
R5 1
e) 3 (x−3)2 dx
77
4.5. Homework exercises
In Problems 10-15, evaluate the improper integral (if the integral diverges, say
so).
10. 13.
Z 1
1
Z 4
1
dx √ dx
0 x2 2 4−x
11. 14. Z ∞
Z 1 1
x ln x dx dx
0 1 x ln x
12. 15.
Z e
2
Z 3
1
ln x dx dx
0 1 x2 −1
In Problems 16-25, determine whether the improper integral converges or di-
verges (you do not necessarily need to evaluate the integral if it converges). Note:
the intent here is for you to use theory, not to perform lots of computations.
16. Z ∞ 21.
5 Z ∞
cos(πx) + 3
dx dx
1 x8 1 x4
17. Z ∞ 22. Z ∞
2 3 2
4
dx − 2 dx
6 x +1 1 x x
18. 23. Z ∞
Z ∞
1 2 5
√ dx 3
+ 4 dx
3 x−1 2 x x
19. 24. Z ∞
Z ∞ sin(2x + 1) + 4
4x −3/2
dx √ dx
2 3 x
20. Z ∞ 25. Z ∞
1 1
dx dx
1 x
e +x 8 2x − 1
78
4.5. Homework exercises
Answers
1
1. 3
5. diverges
2. diverges 6. 31/3
22/3
3. 1
7. Yes, Mathematica recognizes that this
4. π/2 integral diverges.
R ∞ n −x
8. 1; 2; 6; 24; 120. In general, 0 x e dx = n!.
R2 1 R2 1
9. a) 1 x ln x dx = lim+ b x ln x dx
b→1
(Picture looks like the one on page 7, with the asymptote at x = 1)
R1 x Rb x
b) 0 x2 −1 dx = lim− 0 x2 −1 dx
b→1
(Picture looks like the one on page 6, with the asymptote at x = 1)
R∞ 1 R1 1 RB 1
c) 0 xex
dx = lim+ b xex dx + lim 1 xex
dx
b→0 B→∞
(Picture looks like the example on page 10, with the asymptotes at x = 0
and y = 0)
R5 1 Rb 1
d) 3 (x−5)2 dx = lim− 3 (x−5)2 dx
b→5
(Picture looks like the one on page 6, with the asymptote at x = 5)
R5 1 R5 1
e) 3 (x−3)2 dx = lim+ b (x−3)2 dx
b→3
(Picture looks like the one on page 7, with the asymptote at x = 3)
10. diverges
−1
11. 4
79
Chapter 5
5.1 Area
Definite integrals were invented for the purpose of finding areas of regions.
However, there is a slight difference between computing an integral and finding
an area, because integrals can be negative (whereas areas are never negative). For-
mally speaking, integrals compute signed areas.
For example, suppose we are asked to compute the area between the graphs of
f and g between x = a and x = b where f and g are as in the following picture:
This area is
80
5.1. Area
Here is another example. Here f is the solid line and g is the dashed line; the
problem is to compute the total area enclosed by the graphs:
a d b
Example 1: Find the area between the graphs of f (x) = x2 +2 and g(x) = −x−3
on [−2, 3].
81
5.1. Area
Example 2: Find the area enclosed by the graphs of f (x) = 2 − x2 and g(x) = x.
82
5.1. Area
x3 + x2 − 6x = −x3 + 5x2
2x3 − 4x2 − 6x = 0
2x(x2 − 2x − 3) = 0
2x(x + 1)(x − 3) = 0
⇒ x = 0, x = −1, x = 3
83
5.1. Area
Example 4: Find the area of the region which is above the x−axis, below y = 4,
to the left of y = (x − 1)2 and to the right of y = 2x. (The region is pictured below.)
4
-1 1 2 3 4
-1
84
5.1. Area
Example 4 again:
-1 1 2 3 4
-1
85
5.1. Area
86
5.2. Volume
5.2 Volume
Recall that in the previous section we talked about area, which is a 2-dimensional
problem:
a x b
5. “Add up” the areas of these rectangles to get the area of the region:
Z b
A= L(x) dx
a
87
5.2. Volume
5. “Add up” the volunes of these cross-sections to get the volume of the solid:
Z b
V = A(x) dx
a
88
5.2. Volume
Rb
General volume formula: V = a A(x) dx where A(x) = area of the cross-
section at x
Example 1: Consider a solid whose base consists of the region between the y-
axis and the right-half of the circle x2 +y 2 = 16. Cross-sections of the solids parallel
to the y-axis are squares with bases in the xy-plane. Find the volume of the solid.
-1 1 2 3 4 5
-2
-4
89
5.2. Volume
The next example illustrates a method called the disc method (volume), which
is one of three methods used to calculate volumes of solids obtained by revolving
a region around some axis.
Example 2: The region R between the graph of f (x) = x4/3 + 1 and the x−axis
on [0, 2] is revolved around the x−axis to produce a solid. Find the volume of the
solid.
-1 1 2 3
-1
In general, if you take the region between f and the x−axis between x = a and
x = b and revolve that region around the x−axis to produce a solid, the volume of
the solid generated is
Z b
V = π[f (x)]2 dx.
a
This formula is called the disc method because cross-sections are circles/disks.
90
5.2. Volume
Example 3: Find the volume of a solid obtained by revolving the region en-
closed by the graphs of f (x) = x2 and g(x) = x about the line y = 3.
This formula is called the washer method because cross-sections are washer-shaped.
91
5.2. Volume
92
5.2. Volume
10
-1 1 2 3
93
5.2. Volume
√
Example 6: The region enclosed by y = x, y = 2 and x = 0 is revolved around
the line y = −1 to produce a solid. Find its volume.
94
5.2. Volume
95
SOLID COMES FROM REVOLUTION SOLID COMES FROM REVOLUTION
ABOUT A VERTICAL LINE x = L ABOUT A HORIZONTAL LINE y = L
shell method cross-section at x: disk/washer method cross-section at x:
INTEGRATE
WITH
RESPECT
TO x
a x b a x b
(Equations r = |x − L| R = |g(x) − L|
need to be h = |g(x) − f (x)| r = |f (x) − L|
solved for y SA(x) = 2πrh A(x) = πR2 − πr2
in terms
of x)
Rb Rb
⇒V = a 2π |x − L| |g(x) − f (x)| dx ⇒V = a [π(g(x) − L)2 − π(f (x) − L)2 ] dx
disk/washer method cross-section at x: shell method cross-section at x:
INTEGRATE
WITH
RESPECT
TO y
(Equations R = |f −1 (y) − L| r = |y − L|
need to be r = |g −1 (y) − L| h = |f −1 (y) − g −1 (y)|
solved for x A(y) = πR2 − πr2 A(y) = 2πrh
in terms
of y)
Rd Rd
⇒V = c [π(f −1 (y) − L)2 − π(g −1 (y) − L)2 ] dy ⇒V = c 2π|y − L| |f −1 (y) − g −1 (y)| dy
5.3. General principles behind all applications of integration
A = Q · ∆x when Q is constant.
Here are some examples, most of which we have seen before, where this gen-
eral principle is at work:
Rb
volume V cross-sectional V = A · ∆x V = a A(x) dx
area A (x = length)
Rb
work W force F W = F · ∆x W = a F (x) dx
(x = distance)
In the next few sections, we will use this general principle to develop formulas
which give more applications of integration.
96
5.4. Arc length
0 1 2 3 4 5
Remark: in mathematics, the letter s is used for arc length. No one knows why.
Solution: Let’s suppose first that the slope of y = f (x) is constant (so that the
slope is going to be like the Q of the chart on the previous page). That means we
are assuming f (x) is...
Thus
q
s= (∆x)2 + (∆y)2
q
= (∆x)2 + (m∆x)2
q
= (∆x)2 (1 + m2 )
√
= 1 + m2 ∆x.
97
5.4. Arc length
On the previous page, we found that when the slope m is constant, the arc
length from x = a to x = b is
√
s = 1 + m2 ∆x.
That means that if the slope of y = f (x) is nonconstant (and depends on x), the arc
length from x = a to x = b is
s=
(by applying the general principle of integration applications from the previous
section). We have derived the following theorem:
Theorem 5.2 (Arc length formula) Let f be a differentiable function on [a, b]. Then
the length of the graph of y = f (x) from x = a to x = b is
Z bq
s= 1 + [f 0 (x)]2 dx.
a
Example 1: Write an integral which computes the length of the curve y = sin x
from x = 0 to x = π/2.
98
5.4. Arc length
Note: Observe that in order to get an integral that you can actually work out,
you have to start with a weird function. This is your first exposure to the notion
that integrals representing lengths of curves are harder to compute exactly than those
representing areas or volumes of regions. In fact there is an entire branch of math
research devoted to studying integrals which represent the lengths of curves; this
is the theory of elliptic integrals.
99
5.5. One-dimensional moments and centers of mass
The center of mass of the table is the place where you would have to put the
point of the pyramid to balance it.
If the tabletop is circular and made of a uniform material, this is easy. You just
put the center of the table on top of the pyramid. But if the tabletop isn’t round
(or otherwise symmetric) and/or if the tabletop has nonuniform density (i.e. it is
made from a porous material with varying amounts of air pockets in it), it isn’t so
easy to see immediately where the center of mass is. We want to figure out how to
find the center of mass in these situations.
The reason for wanting to know how to compute a center of mass is that in a
physics problem, if you are studying the motion of a large object (like a comet or a
space station orbiting the earth), you can treat the object as a single point (which is
easier to study mathematically than a large blob) so long as the point you use is
the center of mass of the object.
100
5.5. One-dimensional moments and centers of mass
Physics tells us that each mass exerts torque on the system, causing the system
to want to rotate about the fulcrum. Masses on opposite sides of the fulcrum exert
torque in opposite directions.
Adding the torque for each of the masses in the system gives total torque T , where
n
X
T = mj (9.8)(xi − x)
j=1
T =
101
5.5. One-dimensional moments and centers of mass
so if the fulcrum is at x = 8,
T =
If the fulcrum is at x = 6,
T =
If the fulcrum is at x = 7,
T =
When the total torque is zero, we say the system is at equilibrium and the cor-
responding location of the fulcrum is called the center of mass of the system and
is denoted x. (In our example x = 7.)
Now let’s do the same example without using numbers. Suppose individual
masses m1 , m2 , ..., mn at certain points x1 , x2 , ..., xn along a number line. What is x?
We have proven:
102
5.5. One-dimensional moments and centers of mass
Theorem 5.3 (Center of mass for discrete masses along a line) Suppose discrete
masses m1 , m2 , ..., mn are located at respective points x1 , x2 , ..., xn along a number
line. Then:
M0
x= .
M
Question: What is x?
Suppose first that the density δ is constant. Then the total mass is
M=
103
5.5. One-dimensional moments and centers of mass
n
Notice the similarity to the discrete case, where M was mj . Essentially, the
P
j=1
summation is replaced by an integral and the masses mj are replaced by the den-
sity function δ(x). In the same way, the moment about the origin should be
M0 =
Summarizing, we have
Example: Find the center of mass of a thin rod of length 2 units, whose density
x units from the left edge of the rod is δ(x) = x2 + x.
104
5.6. Two-dimensional moments and centers of mass
105
5.6. Two-dimensional moments and centers of mass
We conclude:
Theorem 5.5 (Center of mass for discrete masses in a plane) Suppose that dis-
crete masses m1 , m2 , ..., mn are located at respective points (x1 , y1 ), (x2 , y2 ), ..., (xn , yn )
in a plane. Then:
Setup:
d
a x b
106
5.6. Two-dimensional moments and centers of mass
Question: What is the center of mass (x, y) for a (flat) slab in the plane, possibly
with nonconstant density?
Note: We assume in Math 230 that the density is either constant, or depends
only on x (i.e. the density is δ(x), not δ(x, y)). Otherwise one would need machin-
ery learned in Math 320.
My Mx
x= and y= .
M M
Notice there is no formula for Mx in the above theorem. This is because, when
you generalize the formula for Mx obtained earlier in this section, you get
Z d
Mx = y l(y) δ(y) dy
c
and this is a problem because we are assuming the density depends on x, not y.
107
5.6. Two-dimensional moments and centers of mass
My Mx
x= and y= .
M M
We already have seen statements (1), (3) and (4) in the previous theorem. To
see why (2) is true, think of the formula Mx as a volume computation (which looks
like something obtained by the washer method without the π but with a 12 ). If you
compute the same volume with shells, you get the integral
Z d
y l(y) δ(y) dy
c
108
5.6. Two-dimensional moments and centers of mass
Example: Find the center of mass of the region enclosed by y = 4x − x2 and the
x−axis, where the density function is δ(x) = x + 1.
4
-1 1 2 3 4 5
-1
109
5.6. Two-dimensional moments and centers of mass
Example: Find the center of mass of the region bounded by y = 9 − x2 and the
x−axis, assuming that the density is constant.
10
-3 -2 -1 1 2 3
Notice that this answer did not depend on the actual value of δ. So we may as
well have assumed that the density of the region is 1. This situation arises fairly
often, so we invent the following terminology:
Definition 5.8 The centroid of a region is the center of mass of that region, assuming
it has constant density. (To compute a centroid, just assume δ = 1 and use the formulas
already discussed.)
110
5.6. Two-dimensional moments and centers of mass
Example: Write integrals which can be used to find the centroid of the region
enclosed by y = x and y = x2 − 6.
4
-4 -2 2 4
-2
-4
-6
-8
111
5.7. Probability
5.7 Probability
Finite probability
A motivating example: Suppose you spin a spinner and it stops at a random loca-
tion. You either win or lose money depending on where the spinner lands:
' $
@
@
@ +10
@ +4
−8 @
@A
@Av
A
AU
−2
& %
Definition 5.9 A random variable (r.v) is a quantity X which depends on the re-
sult of some random experiment.
In the spinner example, X is the amount you win/lose (i.e. is the number on
which the spinner lands). To describe X, we need to know two things about it:
1.
2.
This information can be conveyed by means of a table:
x f (x) = P (X = x)
10
-8
-2
112
5.7. Probability
Note: The range of X are the numbers in the left-hand column, and the proba-
bility of each value of X is in the right-hand column.
1/4 v
1/8 v v
- x
−8 −2 4 10
Example: In the context of this spinner, let X be the number spun. What is the
probability that X ≥ 0 (in symbols, we are asking for P (X ≥ 0))?
Solution:
Theorem 5.10 Let X be a random variable taking only the values x1 , x2 , ..., xn , with
respective probabilities f (x1 ), f (x2 ), ..., f (xn ). Then for any set E, the probability that
X is in E is X
P (E) = P (X ∈ E) = f (xj ).
xj ∈E
In other words, for random variables that only take finitely many different val-
ues, probabilities are computed via addition.
113
5.7. Probability
Expected value
A new question: Suppose you spin the spinner 800 times. How much would you
expect to be ahead or behind after the 800 spins?
Definition 5.11 Let X be a random variable taking only the values x1 , x2 , ..., xn , with
respective probabilities f (x1 ), f (x2 ), ..., f (xn ). The expected value of X, denoted
EX, is
n
X
EX = xj f (xj ).
j=1
The expected value of a random variable is the amount you would expect the r.v.
to average if you repeat the experiment over and over.
114
5.7. Probability
In the real world, random variables are not necessarily finite-valued. For exam-
ple, suppose you want to consider the amount of time until some particular event
occurs. Here are some particular examples:
Example: Suppose X is the amount of time until your cell phone rings. What
is the probability that X = 34 minutes?
Example: Suppose X is the amount of time until your cell phone rings. What
is the probability that X ≥ 34 minutes?
115
5.7. Probability
Example: Choose a real number from the interval from the interval [0, 4] “uni-
formly” (this means that all real numbers should be “relatively equally likely" to
be chosen). What is the probability that X ∈ [2, 3]?
This allows us to consider situations where all numbers are not relatively equally
likely. Suppose f : R → R is some function whose graph is like this:
0 2 3 7 8 10 12
The function f above is called a density function for the random variable X.
More generally:
116
5.7. Probability
The definition implies three things that must be true about the density function
of any random variable:
1.
2.
3.
and
Z b
P (X ∈ (a, b]) = P (X ∈ [a, b]) − P (X = a) = P (X ∈ [a, b]) − 0 = f (x) dx
a
117
5.7. Probability
Find the probability that the driver will cause an accident within two months, and
find the expected amount of time until the driver causes an accident.
118
5.7. Probability
where c is a constant.
(a) Find c.
(b) Find P (3 < X < 5).
(c) Find P (X ∈ [−4, 1)).
(d) Find P (X = 7).
(e) Find the expected value of X.
119
5.8. Homework exercises
2. Find the area of the region enclosed by the graphs of f (x) = x2 + 2x + 1 and
g(x) = 2x + 5.
3. Find the area of the region enclosed by the graphs of f (x) = x2 − 4x + 3 and
g(x) = −x2 + 2x + 3.
4. Find the area of the region enclosed by the graphs of f (x) = (x − 1)3 and
g(x) = x − 1.
√
5. Find the area of the region enclosed by the graphs of f (x) = 3x + 1 and
g(x) = x + 1.
6. Find the area of the region lying between the graphs of f (x) = 2 sin x and
g(x) = tan x from x = −π
3
to x = π3 .
7. Find the area of the region lying between the graphs of f (x) = cos x and
g(x) = 2 − cos x from x = 0 to x = 2π.
8. Let R be the region of the xy−plane lying above the x−axis, below the line
1 2
y = 32 x − 4, and above the curve y = 15 x − 32 x + 1. Find the area of R.
10. An ellipse (centered at the origin) is the graph of an equation of the form
2
x2
a2
+ yb2 = 1. Such an ellipse looks like an oval passing through the points
(a, 0), (−a, 0), (0, b) and (0, −b).
a) Write down an expression (in terms of one or more integrals) which will
give the area of this ellipse.
b) Use Mathematica to evaluate the expression you got in part (a).
11. Consider the quadrilateral whose vertices are (in order) (0, 2), (4, 7), (6, 6)
and (2, 0) . Write down an expression involving integrals which gives the
area of this quadrilateral, and evaluate the expression to find the area of the
quadrilateral (you can use Mathematica or do the integrals by hand).
120
5.8. Homework exercises
√
12. Let R be the region enclosed by the graphs of y = x and y = x.
14. Consider a solid whose base is the triangle in the xy−plane whose vertices
are (0, 0), (0, 2) and (6, 0), such that cross-sections of the solid are squares
parallel to the y−axis. Find the volume of the solid.
15. Consider a solid whose base is the region in the xy−plane lying above y = x
and below y = 2 − x2 , such that cross-sections of the solid parallel to the
y−axis are semicircles whose diameter lies in the xy−plane. Find the volume
of the solid.
16. Consider a solid whose base is the interior of the circle x2 + y 2 = 16 in the
xy−plane. If cross-sections of the solid parallel to the y−axis are rectangles
which are twice as high as they are wide, find the volume of the solid.
17. Find the volume of the solid obtained by revolving the region below y = sec x
and above the x−axis from x = 0 to x = π4 around the x−axis.
18. Find the volume of the solid obtained by revolving the region below y =
2 − x2 and above y = 0 around the x−axis.
In Problems
√ 19-22, let R be the region in the xy−plane bounded by the graphs
y = x, y = 0 and x = 4.
121
5.8. Homework exercises
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S1 .
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S1 .
c) Evaluate the integrals from parts (a) and (b) using Mathematica, and ver-
ify that you get the same answer.
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S2 .
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S2 .
c) Evaluate the integrals from parts (a) and (b) using Mathematica, and ver-
ify that you get the same answer.
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S3 .
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S3 .
22. Let S4 be the solid obtained by revolving R around the line y = −3.
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S4 .
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S4 .
In Problems 23-35, let Q be the region in the xy−plane lying to the right of the
y−axis, above the curve y = 2x2 and below the line y = 8.
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S1 .
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S1 .
24. Let S2 be the solid obtained by revolving R around the line x = −4.
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S3 .
122
5.8. Homework exercises
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S3 .
a) Write a formula involving one or more integrals with respect to the vari-
able x which gives the volume of S4 .
b) Write a formula involving one or more integrals with respect to the vari-
able y which gives the volume of S4 .
26. Use calculus to show that the volume of a sphere of radius r is 34 πr3 . Hint:
You can visualize such a sphere by first thinking of a circle of radius r in
the plane, and building a solid over the base where the cross-sections of the
solid parallel to the y−axis are themselves circles. Write down an integral
which gives the volume, and evaluate the integral by hand (it’s not hard) to
get 34 πr3 .
27. Suppose a bird flies in a straight line so that its velocity at time t is 50t2 − 20t
mi/hr. Find the distance the bird travels between times t = 0 and t = 1.
28. If the density of a metal rod is constant, then the mass of the metal rod is
equal to the density times the length of the rod. Suppose a metal rod has a
nonconstant density, where the density x units from the left edge of the rod
is 2x2 + x + 1 grams/unit. Find the mass of the rod, if the wire is 3 units long.
29. If the force applied to an object is constant, then the work done by that force
is equal to the force times the distance the object moves. Suppose a force is
applied to an object where the force at position x is 2x + 3 Newtons. Find the
work done in moving the object from position 5 meters to position 10 meters.
30. If the current i in an electric circuit is constant, then the charge q that builds
up in the circuit is equal to the current times the amount of time that the
current is in the circuit. Find the charge in an electric circuit if a current of
sin t + 2 cos t + 3 is applied to the circuit from time 0 to time π.
31. Find (by hand) the length of the curve y = 23 x3/2 + 1 from x = 0 to x = 1.
32. Find (by hand) the length of the curve y = 2x3/2 − 2 from x = 0 to x = 8.
33. Write an integral which gives the length of the curve y = ln(sin x) between
x = π4 and x = π2 , and use Mathematica to evaluate the integral.
√3
34. Write an integral which gives the length of the curve y = x2 between the
points (1, 1) and (8, 4), and use Mathematica to evaluate the integral.
123
5.8. Homework exercises
35. Write an integral which gives the length of the curve y = x1 from x = 1 to
x = 3, and use Mathematica to find a decimal approximation to this integral.
36. Write an integral which gives the length of the curve y = sin x from x = 0 to
x = π, and use Mathematica to find a decimal approximation to this integral.
37. Find the total length (you can use Mathematica to do the integral) of the curve
x2/3 + y 2/3 = 4 (the graph of this curve is found below (it extends 8 units from
the origin to the left, right, up and down):
-8 8
-8
38. Suppose a system consists of four objects, with respective masses 7, 4, 3 and
8, located along an axis with respective positions −3, −2, 5 and 6. Find the
total mass of the system, the moment about the origin of the system, and the
center of mass of the system.
39. Suppose a system consists of three objects located along an axis. The masses
of the three objects are 6, 5 and 3, and the respective positions of the objects
are −5, 1 and 3.
40. Suppose two siblings sit on the ends of a see-saw; a brother, who weighs 75
pounds, and a sister, who weighs 50 pounds. If the seesaw is 12 feet long,
how far from the brother should the fulcrum be placed so that the see-saw is
balanced?
41. A metal rod of length 10cm has density δ(x) = 1 + x2 g/cm at a distance x cm
from the left end of the rod. Find the mass of the rod, and find how far from
the left end of the rod the center of mass is.
42. A rod with density δ(x) = 2 + sin x kg/unit is positioned along the positive
x−axis, with its left end at x = 0 and its right end at x = 3π 4
. Find the
x−coordinate of the center of mass of the rod.
43. Suppose a system consists of three objects in a plane. The first mass is 5g,
located at (2, 2); the second mass is 1g, located at (−3, 1), and the last mass is
3g, located at (1, −4). Find the total mass of the system, the moments about
the x− and y− axis of the system, and the center of mass of the system.
44. Find the center of mass of a system which consists of the following five
masses:
124
5.8. Homework exercises
mass 3 4 2 1 6
position (−2, −3) (5, 5) (7, 1) (0, 0) (−3, 0)
√
45. Let R be the region in the xy−plane above the x−axis, below y = x and to
the left of x = 9.
46. Let R be the region in the first quadrant between the graphs of y = x2 and
y = x3 .
47. Find the centroid of the region bounded by the graphs of y = xe−x , y = 0 and
x = 5. (Use Mathematica to evaluate the integrals.)
48. Find the centroid of a semicircle of radius r (a picture is shown below, but
you have to figure out what graph is the top of the semicircle, and how far to
the left or right it goes).
49. Find the centroid of a circle of radius r centered at the origin. Hint: there is a
more clever approach than doing a bunch of integrals.
50. Consider the region in the xy−plane bounded by the graphs of y = x2 and
y = b, where b > 0 is a positive constant. What is the x−coordinate of the
center of mass of this region? Is the y−coordinate of the center of mass of this
region greater than, or less than 2b ? Explain your reasoning.
51. Let X be a random variable taking values in the interval [0, 1]. If the density
function of X is f (x) = cx2 + x, find the value of c.
52. Let X be a random variable taking values in the interval [0, 2]. If the density
function of X is f (x) = ce−x/2 , find the value of c.
125
5.8. Homework exercises
53. Let X be a random variable taking values in [0, 4] whose density function is
f (x) = 18 x.
54. Let X be a random variable taking values in [0, ∞) whose density function is
f (x) = ce−3x .
a) Find c.
b) Find the probability that X > 10.
c) Find the probability that 3 ≤ X ≤ 4.
55. Suppose that the lifespan of a certain organism is a random variable X whose
density function is f (x) = 4xe−2x (where x ∈ (0, ∞) is measured in years).
a) Find the probability that the organism lives for less than one year.
b) Find the expected lifespan of the organism.
56. Suppose that the lifetime (in months) of a light bulb is a random variable X
1
with density function f (x) = 81 e− 8 x (again, X takes values in (0, ∞).
a) Find the probability that the light bulb lasts more than four months.
b) Find the expected lifetime of the lightbulb.
57. Find the expected value of a random variable whose density function is f (x) =
1
2
sin x for 0 ≤ x ≤ π.
58. Let X be a continuous random variable taking values in some interval. The
median of X is a number m such that P (X ≤ m) = 21 . In each of these
examples, find the median of X, if X has the given density function:
126
5.8. Homework exercises
Answers
3
1. 36 5. 2
32
2. 3
6. 2 − ln 4
3. 9 7. 4π
√
1 103−4 10
4. 2
8. 9
9. a) x2 + y 2 = r2
R √
b) A = 4 0r r2 − x2 dx
c) πr2
d) πr2
Ra q x2
10. a) A = 4 0 b 1− a2
dx
b) πab
R2h 5 i R4h 5 i
11. A = ( x + 2) − (2 − x)
0 4
dx + 2 ( x + 2) − ( 3 x − 3)
4 2
dx
R h i
+ 6 ( −1 x + 9) − ( 3 x − 3)
4 2 2
dx = 18
√
12. a) The curved graph on the top is y = x; the diagonal line is y = x:
1.0
0.8
0.6
0.4
0.2
R1 √ R1 1
b) 0 ( x − x) dx c) 0 (y − y 2 ) dy d) 6
13. a) The line on the left is x = 0, the line in the upper right is 3x + y = 6 and
the curve is y = x2 − 4.
6
-2
-4
R2 2 34
b) 0 [(6 − 3x) − (x − 4)] dx d) 3
R0 √
y + 4 dy + 06 (2 − y3 ) dy
R
c) −4
127
5.8. Homework exercises
R6
14. V = 0 [2 − 13 x]2 dx = 8
R 1 1 2−x2 −x 2 81
15. V = −2 2π 2
dx = 80
π
R4 √ 2
2048
16. V = −4 8 16 − x2 dx = 3
R π/4
17. V = 0 π sec2 x dx = π
R √2 √
18. V = √ π(2 − x2 )2 dx = 64 2 π
− 2 15
R4
19. a) V = 0 πx dx
R2
b) V = 0 2πy(4 − y 2 ) dy
c) 8π
R4 √
20. a) V = 0 2πx x dx
R2
b) V = 0 [π42 − π(y 2 )2 ] dy
128
c) 5
π
R4 √
21. a) V = 0 2π(6 − x) x dx
R2
b) V = [π(6 − y 2 )2 − π(6 − 4)2 ] dy
0
√
a) V = 04 [π( x + 3)2 − π(0 + 3)2 ] dx
R
22.
R2
b) V = 0 2π(y + 3)[4 − y 2 ] dy
R2
23. a) V = 0 [π(8)2 − π(2x2 )2 ] dx
R8 q
y
b) V = 0 2πy 2
dy
R2
24. a) V = 0 2π(x + 4)[8 − 2x2 ] dx
R8h q
y
i
b) V = 0 π( 2
+ 4)2 − π(0 + 4)2 dy
R2
25. a) V = 0 π(8 − 2x2 )2 dx
R8 q
y
b) V =0 2π(8 − y)[ 2 ] dy
Rr √
26. V = −r π( r2 − x2 )2 dx = 43 πr3
20
27. 3
mi
51
28. 2
g
29. 90 Nm
30. 2 + 3π
128
5.8. Homework exercises
−2+2(2)3/2
31. 3
−2+2(73)3/2
32. 27
R π/2 √
33. s = π/4 1 + cot2 x dx = − ln(tan π8 )
√ √
r h i2
R8 1 −1/2 80 10−13 13
34. s = 1 1+ 2
x dx = 27
R3q
35. s = 1 + [−x−2 ]2 dx ≈ 2.14662
1
R √
36. s = 0π 1 + cos2 x dx ≈ 3.8202
37. 48
I have not found the answers to problems 38-50 on moments and centers of
mass.
3
51. c = 2
e
52. c = 2e−2
3
53. a) P (X > 2) = 4
b) P (X = 3) = 0
1
c) P (X ≤ 1) = 16
d) They are equally likely (the probability of both is zero).
e) It is more likely that X is very close to 3 since f (3) > f (1).
f) b = √4 .
3
54. a) c = 3
b) P (X > 10) = e−30
c) P (3 ≤ X ≤ 4) = e−9 − e−12 .
55. a) P (X < 1) = 1 − 3e−2
b) EX = 1 year
56. a) P (X > 4) = e−1/2
b) EX = 8 months
π
57. EX = 2
58. a) m = 2−1/4
b) m = 21 ln 2
129
Chapter 6
Parametric equations
y = f (x)
-3 -2 -1 1 2 3
-1
-2
-3
130
6.1. Introduction to parametric equations
One thing we know is that the graph of any function y = f (x) must pass the
Vertical Line Test: any vertical line must hit the graph of y = f (x) in at most one
point.
10 10
5 5
-10 -5 5 10
-6 -4 -2 2 4
-5 -5
-10 -10
Suppose we consider some curve which does not pass the Vertical Line Test. Is
there any way to represent such a curve as a function (and subsequently compute
derivatives, integrals and their applications)?
The answer is yes; the method is to think of both the x− and y− coordinates of
functions of a third variable, which is usually called t. The resulting set of equa-
tions (
x = f (t)
y = g(t)
are called parametric equations for the curve. As an example, consider the set of
parametric equations (
x = t2 − 1
y = t − t2
To obtain a graph of this set of equations, choose some values of t and solve for the
corresponding x and y. Plot the points x and y to obtain the graph. In our example,
we have
t -2 -1 0 1 2
x = f (t)
y = g(t)
(x, y)
131
6.1. Introduction to parametric equations
-5 -4 -3 -2 -1 1 2 3 4 5
-1
-2
-3
-4
-5
-6
-7
-8
Note: For each t, the point (f (t), g(t)) should be thought of as the location of an
object at time t, and the curve is the path the object traces out as it moves. Thus
the graph of a set of parametric equations indicates a direction of motion along the
curve (from negative t to positive t) as well as the curve itself.
Our goal in this chapter is to analyze the motion of an object whose position at
time t is given by parametric equations like these. In particular, we would like to
know:
132
6.1. Introduction to parametric equations
133
6.2. Parametric equations of common graphs
(x − h)2 + (y − k)2 = r2 .
-1 1 2 3 4 5 6 7
-1
-2
-3
-4
-5
-6
Theorem 6.1 (Parametric equations of a circle) The circle centered at (h, k) with
radius r, whose Cartesian equation is (x−h)2 +(y−k)2 = r2 , has parametric equations
(
x = r cos t + h
.
y = r sin t + k
134
6.2. Parametric equations of common graphs
Example: Find parametric equations for the circle (x + 2)2 + y 2 = 7, where the
circle is traced out counterclockwise.
Question: What if you want the circle to be traced out clockwise? (Later)
Ellipses
An ellipse is the graph of any Cartesian equation of the form
(x − h)2 (y − k)2
+ = 1.
a2 b2
(If a = b, then the ellipse is actually a circle, so circles are in fact special kinds of
ellipses.)
135
6.2. Parametric equations of common graphs
As with circles, these parametric equations trace out the ellipse counterclock-
wise, with t = 0 corresponding to the right-most point on the circle and t generally
corresponding to the angle of the point on the ellipse, measured from the positive
x−direction.
Example: Sketch the graph of the ellipse whose parametric equations are x =
2 cos t − 1, y = 3 sin t + 5.
10
-6 -5 -4 -3 -2 -1 1 2 3 4 5
-1
136
6.2. Parametric equations of common graphs
Hyperbolas
A hyperbola is the graph of any Cartesian equation of the form
To obtain parametric equations for a hyperbola, one mimics the procedure that
was done for ellipses. The difference is that to obtain the minus sign in the equa-
tion, one starts not by thinking of
cos2 t + sin2 t = 1
but by thinking of
sec2 t − tan2 t = 1.
This eventually yields the following:
137
6.2. Parametric equations of common graphs
Remark: There are other parametric equations for hyperbolas, which we do not
mention here.
Remark: I am not mentioning the direction the hyperbolas are traced out by
these parametric equations here. Ask me outside of class if you want to know how
this works.
Example: Sketch the graph of the hyperbola whose parametric equations are
x = sec t + 2, y = 4 tan t.
6
-6 -5 -4 -3 -2 -1 1 2 3 4 5 6
-1
-2
-3
-4
-5
Example: Sketch the graph of the hyperbola whose parametric equations are
x = 3 tan t − 1, y = sec t.
6
-6 -5 -4 -3 -2 -1 1 2 3 4 5 6
-1
-2
-3
-4
-5
138
6.2. Parametric equations of common graphs
To summarize:
Theorem 6.5 (Parametric equations of a line) The parametric equations of the line
passing through points (x0 , y0 ) and (x1 , y1 ) are
(
x = x0 + (x1 − x0 )t
y = y0 + (y1 − y0 )t
These parametric equations trace out the line in the direction moving from
(x0 , y0 ) to (x1 , y1 ), with t = 0 corresponding to (x0 , y0 ) and t = 1 corresponding
to (x1 , y1 ).
139
6.2. Parametric equations of common graphs
the segment between (x0 , y0 ) and (x1 , y1 ), use the same parametric equations as the
line through those points and truncate the values of t:
These parametric equations trace out the segment in the direction moving from
(x0 , y0 ) to (x1 , y1 ), with t = 0 corresponding to (x0 , y0 ) and t = 1 corresponding to
(x1 , y1 ).
Example: Find parametric equations for the line x = −2, which trace out the
line downward.
Example: Find parametric equations for the segment connecting the points
(−3, 2) and (4, −5).
140
6.2. Parametric equations of common graphs
Example: Sketch the graph of the curve whose parametric equations are
(
x=3−t
, where t ranges from 0 to 1.
y = −2 + 3t
6
-6 -5 -4 -3 -2 -1 1 2 3 4 5 6
-1
-2
-3
-4
-5
These parametric equations trace out the curve y = h(x) from left to right,
where the value of t is the same as the value of x.
141
6.2. Parametric equations of common graphs
-6 -5 -4 -3 -2 -1 1 2 3 4 5 6
-1
-2
-3
-4
-5
x = r cos t + h
Circle center (h, k) radius r
y = r sin t + k
x = a cos t + h
Ellipse center (h, k)
y = b sin t + k
x = a sec t + h
Hyperbola center (h, k) opening left/right
y = b tan t + k
x = a tan t + h
Hyperbola center (h, k) opening up/down
y = b sec t + k
x = x0 + (x1 − x0 )t
Line through (x0 , y0 ) and (x1 , y1 )
y = y0 + (y1 − y0 )t
x=t
Cartesian function y = h(x)
y = h(t)
142
6.3. Calculus with parametric equations
Question: How do you describe the velocity, speed, acceleration of the object?
To describe the physical motion of such an object, we need the notion of a two-
dimensional vector. The idea of a vector is made more precise in linear algebra,
but for now you should think of a two-dimensional vector as a mathematical ob-
ject sitting in an xy−plane which has two (and only two) attributes: a length and
a direction. (In linear algebra you will learn that vectors are actually much more
abstract than the ones we think about here.)
143
6.3. Calculus with parametric equations
Note: A vector is really more like a “floating” arrow in that what is relevant
is how far to the left/right and up/down the vector points, not where the vector
starts and ends. For example, the two vectors
are the same vector since they point in the same direction and have the same
length. In practice, the place where you locate a vector depends on the context
of the problem under consideration.
When writing by hand, you usually put an arrow over a letter to emphasize
that the letter represents a vector rather than a number; when typed, a letter is put
in boldface to indicate that it is a vector.
The length of a vector comes from the Pythagorean Theorem / Distance √ 2 For-
mula: given vector v = → −
v = (x, y), the length of v, denoted ||v||, is x + y 2 . In
q √ √
particular, the length of (3, −4) is ||(3, −4)|| = 32 + (−4)2 = 9 + 16 = 25 = 5.
144
6.3. Calculus with parametric equations
Definition 6.8 Suppose an object is moving in an xy−plane such that its position at
time t is the point/vector p = (x, y) where
(
x = f (t)
.
y = g(t)
The speed of the object at time t is the length of the velocity vector:
q
s(t) = [f 0 (t)]2 + [g 0 (t)]2 .
Note: The velocity of an object is a vector (it indicates a speed and a direction)
and should be drawn such that the tail of the vector is at the point (f (t), g(t)). The
velocity vector will always point in a direction tangent to the position curve, and
point in the same direction the curve is being traced out.
However, the speed of an object is a scalar (a.k.a. number). The speed is the
length of the velocity vector; the longer the velocity vector is, the faster the object
is moving.
The acceleration of an object is also a vector which should be drawn such that
the tail of the vector is at the point (f (t), g(t)). The acceleration vector will always
point to the inside of the position curve, looking forward in time from (f (t), g(t)).
(Another way of saying this is that the curve will “bend” towards the acceleration
vector.)
145
6.3. Calculus with parametric equations
Example: Suppose a bug is crawling around an xy−plane such that its position
at time t is (x, y) where x = 2t3 + t and y = 1 − t + t2 − t3 . Find the velocity and
acceleration of the particle at time t = 1.
Example: Suppose at time t = 6 the position vector of a particle is (5, 2), the
velocity vector of the particle is (−2, 3) and the acceleration vector of the particle
is (−4, 1). Find the speed of the particle at time t = 6 and sketch a graph of the
particles motion at times close to t = 6, indicating the velocity and acceleration
vectors appropriately.
6
-6 -5 -4 -3 -2 -1 1 2 3 4 5 6
-1
-2
-3
-4
-5
146
6.3. Calculus with parametric equations
Between two times t = a and t = b, the object will follow some path. A natural
question to ask is what the length of this path. Equivalently, we want to know how
far the object travels between times t = a and t = b.
Suppose for now that the speed of the object is constant. Then the distance the
object travels over a length ∆t of time is
By the general principle of application of integrals, that means that if the speed
is nonconstant, the distance travelled by the object between times t = a and t = b
is
Theorem 6.9 (Arc length formula) If x = f (t); y = g(t) are the parametric equa-
tions of some curve, then the length of that curve between times t = a and t = b
is v
u dx 2
Z bu ! !2
dy
s= t
+ dt.
a dt dt
Example: Use the arc length formula to verify that the circumference of a circle
of radius r is 2πr.
147
6.3. Calculus with parametric equations
and you want to know the slope of the graph of these parametric equations at some
value of t. We know from Calculus I that the slope is given by “the” derivative.
But now there are lots of things which are derivatives:
we have
dy
But what is the derivative dx
?
dy dy dx
= ·
dt dx dt
dy
Solving for dx
, we obtain
148
6.3. Calculus with parametric equations
is
dy
dy dt g 0 (t)
= dx = .
dx dt
f 0 (t)
when t = π4 .
Higher-order derivatives
In the previous section we discussed how to find the slope of a curve whose para-
metric equations are given. Now, we discuss how to find the concavity of such a
curve. The concavity is given by
d2 y
,
dx2
the second derivative of y with respect to x.
d2 y
Example: Find dx2
for the curve whose parametric equations are
(
x = t2 − 1
y = t2 + t
149
6.3. Calculus with parametric equations
To fix this, we again use the Chain Rule. We know (by the Chain Rule) that for
any quantity A,
d dA dA dx
(A) = = ·
dt dt dx dt
so we obtain
dA d
dA dt
(A)
= dx = dt dx .
dx dt dt
dy
Substituting dx
in for A in the above equation, we obtain
! d dy
d2 y d dy dt dx
= = dx .
dx2 dx dx dt
We have established:
150
6.3. Calculus with parametric equations
is
2 d dy
dy dt dx
= dx .
d2 x dt
d2 y
Earlier example, revisited: Find dx2
for the curve whose parametric equations
are (
x = t2 − 1
y = t2 + t
151
6.3. Calculus with parametric equations
q
Speed (length of the velocity vector) [f 0 (t)]2 + [g 0 (t)]2
Rbq
Arc length from t = a to t = b s= a [f 0 (t)]2 + [g 0 (t)]2 dt
dy
dy g 0 (t)
Slope of tangent line dx
= dt
dx = f 0 (t)
dt
d dy
d2 y dt ( dx )
Concavity dx2
= dx
dt
Position, velocity and acceleration are vector quantities; the others are numbers
(a.k.a. scalars).
152
6.4. Transformations on parametric equations
• If k > 0, the equations (#) trace the curve out in the same direction as (∗).
• If k < 0, the equations (#) trace the curve out in the opposite direction as (∗).
• If |k| < 1, the equations (#) trace the curve out more slowly than (∗); the
closer |k| is to zero, the more slowly the curve is traced out.
• If |k| > 1, the equations (#) trace the curve out more quickly than (∗); the
larger |k| is, the faster the curve is traced out.
• If |k| = 1, the equations (#) trace the curve out at the same rate as (∗) (in the
same direction if k = 1; in the opposite direction if k = −1).
153
6.4. Transformations on parametric equations
154
6.5. Homework exercises
1. y = 2x − 3 6. y = x2 − x + 1
12. The line passing through (3, −1) and (−2, −3).
13. The ellipse passing through (−2, −6), (−2, 10), (0, 2) and (−4, 2).
14. The ellipse centered at (4, −1) which passes through (12, −1) and (4, −4).
16. Write parametric equations which trace out the circle (x − 3)2 + (y + 4)2 = 25
in a counterclockwise direction.
(x−1)2 2
17. Write parametric equations which trace out the ellipse 6
+ (y+3)
36
= 1 in a
counterclockwise direction.
(y−2)2
18. Write parametric equations which trace out the ellipse x2 + 16
= 1 in a
counterclockwise direction.
155
6.5. Homework exercises
29. x = 17 t + 72 , y = 74 t − 3
7
In problems 30 to 35, assume that an object’s position (x, y) is described by the set
of parametric equations. Find the velocity and acceleration of the object at time t.
√
30. x = 2t2 + 3, y = t sin 2t 33. x = t2 + 1, y = 2t
36. Suppose an insect is crawling around a plane so that its position at time t is
given by x = 2t + 1 + 2et , y = 7t − 3 − 4et . Find the speed of the insect at time
t, and the speed of the insect at time 1.
37. Suppose a planet is orbiting a star in such a way that its position in some
plane at time t is x = r cos t + h, y = r sin t + k. Show that the planet’s speed
is constant.
38. Suppose an ant is crawling around on an xy−plane so that its position at time
t is given by x = t + sin πt, y = t3 − 2t.
a) What is the position of the ant at time t = 1?
b) What is the ant’s velocity at t = 1?
c) What is the ant’s speed at t = 1?
d) What is the ant’s acceleration at t = 1?
e) Sketch a picture of the ant’s motion at times near t = 1, indicating the
velocity and acceleration vectors appropriately.
f) Write an integral which will compute the distance the ant crawls be-
tween times t = 1 and t = 3. Use Mathematica to compute a decimal
approximation to this integral.
39. Suppose an airplane is flying (at constant height) so that its position on an
xy−grid at time t is (x, y), where x = t sin πt and y = t cos πt.
a) Find a point in the plane that the plane flies over twice. At what times
does the plane fly over this point?
b) Find the position of the airplane at time t = 1.
c) Find the velocity of the airplane when t = 1.
d) Find the speed of the airplane when t = 1.
e) Sketch a picture of the plane’s motion at times near t = 1, indicating the
velocity and acceleration vectors appropriately.
156
6.5. Homework exercises
f) Write an integral which will compute the distance the plane flies be-
tween times t = 1 and t = 2. Use Mathematica to compute a decimal
approximation to this integral.
40. Use Mathematica to find a decimal approximation to the length of the curve
whose parametric equations are x = ln t, y = t + 1 from t = 1 to t = 6.
41. Find the exact length of the curve whose parametric equations are x = e−t cos t,
y = e−t sin t from t = 0 to t = π2 .
42. Write an integral which will find the length of a curve whose parametric
2
equations are x = t −1
t
, y = 3t cos t from t = 1 to t = 2. Use Mathematica to
find a decimal approximation to this integral.
dy
43. Let x = 2 + sec t and y = 1 + 2 tan t. Find dx
in terms of t.
dy d2 y
44. Let x = 2 cot t and y = 2 sin2 t. Find dx
and dx2
in terms of t.
45. Find√the slope
√ of the line tangent to the graph of the parametric equations
x = t, y = t − 1 when t = 2.
46. Find the equation of the line tangent to the graph of the parametric equations
x = cos3 t, y = sin3 t when t = π4 .
47. Consider the graph of the parametric equations x = cos t + t sin t, y = sin t −
t cos t.
a) Find all t in [0, 2π] where the line tangent to the graph is horizontal.
b) Find all t in [0, 2π] where the line tangent to the graph is vertical.
48. Consider the graph of the parametric equations x = 1 − t, y = t3 − 3t.
a) Find all t where the line tangent to the graph is horizontal.
b) Find all t where the line tangent to the graph is vertical.
dy d2 y
49. Suppose x = 2t and y = 3t − 1. Find dx
and dx2
for arbitrary t, and when
t = 3.
dy d2 y
50. Suppose x = 2 cos t and y = 2 sin t. Find dx
and dx2
for arbitrary t, and when
t = π3 .
51. Write parametric equations which trace out the circle (x − 1)2 + (y + 8)2 = 9
in a clockwise direction.
2
52. Write two sets of parametric equations which trace out the ellipse (x+4)
18
+
(y−1)2
36
= 1 in a counterclockwise direction, where the speed of the second set
of equations is twice that of the first.
157
6.5. Homework exercises
2
53. Write parametric equations which trace out the ellipse (x + 3)2 + (y−1)
49
= 1 in
a clockwise direction.
54. Suppose the graph of some set of unknown parametric equations x = f (t),
y = g(t) is given below (the direction of motion is from the upper right to the
lower right):
6
-6 -5 -4 -3 -2 -1 1 2 3 4 5 6
-1
-2
-3
-4
-5
-6
Sketch the graphs of each of the following sets of parametric equations, in-
cluding the direction of motion:
a) x = f (2t),y = g(2t)
b) x = f (−t), y = g(−t)
c) x = f (t) − 2, y = g(t)
d) x = f (t) + 1, y = g(t) − 3
e) x = 2f (t), y = g(t) − 1
f) x = −f (t), y = g(t)
g) x = −2f (t), y = −g(t)
h) x = 21 f (t) − 2, y = 2g(t) + 1
i) x = f (t + 3), y = g(t + 3)
j) x = g(t), y = f (t)
158
6.5. Homework exercises
Answers
33. →
−
v (t) = (t(t2 + 1)−1/2 , 2);
→
−
a (t) = ((t2 + 1)−1/2 , −t2 (t2 + 1)−3/2 , 0)
34. →
−
v (t) = (−6e−3 , 2t − 1);
→
−
a (t) = (18e−4 , 2)
35. →
−
v (t) = (8 sec2 2t, −12 csc2 4t)
→
−
a (t) = (32 sec2 2t sec 2t, 96 csc2 4t cot 4t)
159
6.5. Homework exercises
q
36. s = (2 + 2et )2 + (7 − 4et )2 ; s(1) = 5
39. a) The plane flies over the point (1/2, 0) at times t = 1/2 and t = −1/2.
b) (0, −1).
c) (−π, −1)
√
d) π 2 + 1
e) Picture omitted
R2q
f) 1 [πt cos πt + sin πt]2 + [cos πt − πt sin πt]2 dt ≈ 4.82124
40. 5.38402
√
41. 2(1 − e−π/2 )
rh i2
t2 +1
+ [3 cos x − 3x sin x]2 dt ≈ 4.4799
R2
42. s = 1 t2
dy
43. dx
= 2 csc t
dy d2 y
44. dx
= −4 cos 2t sin2 t sin 2t; dx2
= 2 sin4 t(1 + 2 cos 2t + 3 cos 4t)
√
45. 2
1 1
46. y = −(x − √
2 2
) + √
2 2
47. a) t = π, t = 2π
b) t = π2 , t = 3π
2
48. a) t = −1, t = 1
b) There is no such t.
dy dy d2 y d2 y
49. dx
= 32 ; dx t=3
= 32 ; dx2
= 0; dx2 t=3
=0
dy dy −1 d2 y −1 d2 y
50. dx
= − cot t; dx t=π/3
= √ ;
3 dx2
= 2
csc3 t; dx2 t=π/3
= −4.
160
6.5. Homework exercises
161
Chapter 7
Here is an argument that attempts to demonstrate that I will win this race:
162
7.1. Motivation and big-picture questions
Here is an argument which attempts to demonstrate that I will not win the race:
Examples:
1 + 1 + 1 + 1 + 1 + 1 + ...
1 + 12 + 41 + 18 + 1
16
+ ...
163
7.1. Motivation and big-picture questions
1 + 12 + 31 + 14 + 15 + 61 + ....
1 − 21 + 13 − 14 + 51 − 16 + ...
An interesting example:
1 − 1 + 1 − 1 + 1 − 1 + 1 − 1 + 1 − 1 + ...
Solution #1:
Solution #2:
Solution #3:
This interesting example illustrates a major problem with trying to study infi-
nite series. Irrespective of which of these solutions is correct, what we know is that
they cannot all be correct. This means that we definitely cannot legally rearrange
or regroup terms when adding up infinitely many numbers.
164
7.1. Motivation and big-picture questions
3+8+4+7+2+9
The reason this works is that you can add any two numbers at a time and get
the same answer (this is called the associative property). For example,
3+8+4+7+2+9
But, with an infinite list of numbers there are two problems with this approach:
1.
2.
165
7.2. Convergence and divergence
1. when adding the numbers two at a time, you never run out of numbers to
add; and
Find the
slope of the
line tangent
to function
f at x
Find the
area under
the graph of
function f
from a to b
Find the
sum of
an infinite
list of numbers
a1 + a2 + a3 + ...
166
7.2. Convergence and divergence
Definition 7.1 Given an infinite series a1 + a2 + a3 + a4 + ..., and given any index
N , the N th partial sum of the series, denoted SN , is
SN = a1 + a2 + a3 + ... + aN .
Remark: The indexing of an infinite series does not always start with the index
1. In general, the N th partial sum of an infinite series is the sum of all terms in the
series whose index is ≤ N . For example, if your series is
a5 + a6 + a7 + a8 + a9 + ...,
In particular, to get SN , you add up terms until you get to index N (you don’t
necessarily add up N terms).
Example: Find the second, fourth and fifth partial sums of each of the following
infinite series (assume that each series starts with a1 ):
1 − 1 + 1 − 1 + 1 − 1 + 1 − 1 + ...
1 1 1 1 1
1− + − + − + ...
2 3 4 5 6
Solution:
1 1
S2 = 1 − = .
2 2
1 1 1 7
S4 = 1 − + − = .
2 3 4 12
1 1 1 1 47
S5 = 1 − + − + = .
2 3 4 5 60
167
7.2. Convergence and divergence
1. If L is a real number such that lim SN = L, then we say that the infinite series
N →∞
a1 + a2 + a3 + ... converges (to L) and write a1 + a2 + a3 + ... = L. In this
setting L is called the sum of the series.
168
7.3. Σ-notation
7.3 Σ-notation
Writing a1 + a2 + a3 + ... over and over again is annoying. We shorthand this
expression by writing
∞
X
an
n=1
In this notation:
• n is the
• 1 is the
• ∞ is the
Example:
∞
X n
n=1 n + 3
Note: We will see that some of the things we want to say about series do not
depend on the starting index of the series. In this setting, we will just write an
P
to represent the series. If you see an (without the upper and lower indices indi-
P
1. The starting index was given earlier in the problem and is being omitted
solely for the sake of brevity (while this is “legal”, it is not recommended
that you do this), or
2. Some property of the series is being described which does not depend on the
starting index of the series (so the starting index is irrelevant to the context).
169
7.3. Σ-notation
Example: For each series, write the series out with + signs, and give the second
term, the third partial sum, and the ninth term of the series:
∞
X 1
n=1 n
∞
X (−1)n
2
n=3 n + 2
∞
X
n
n=0
170
7.3. Σ-notation
2 2 2 2 2
(b) + + + + +...
3 7 11 15 19
7 10 13 16 19
(c) + + + + +...
8 16 25 26 27
1 1 1 1 1
(d) 1− + − + − +...
5 9 13 17 21
171
7.4. Elementary properties of convergence and divergence
Definition 7.3 Let an be an infinite series. For each N , let SN be the N th partial
P
sum of the series; this is defined to be the sum of all the an for which n ≤ N . Then:
1. If L is a real number such that lim SN = L, then we say the infinite series
N →∞
a1 + a2 + a3 + ... converges (to L) and write an = L. In this setting L is
P
In particular, you should never omit the Σ when describing whether or not
an infinite series converges.
Now, we list some elementary results about convergence of series. They should
remind you of similar results for improper integrals (there is a good reason for this,
as we will see later).
172
7.4. Elementary properties of convergence and divergence
Theorem 7.6 (Linearity III) Suppose an is an infinite series that diverges and
P
∞
Theorem 7.7 (Starting Index is Irrelevant) Suppose an is an infinite series.
P
n=K
Then, so long as all the terms are defined, for any constant M we have:
∞ ∞
1. an converges if and only if an converges.
P P
n=M n=K
∞ ∞
2. an diverges if and only if an diverges.
P P
n=M n=K
∞
Note: Suppose that an converges. Then since starting index is irrelevant,
P
n=M
∞
then an converges as well. But in general, these two series do not converge to
P
n=K
the same sum.
173
7.4. Elementary properties of convergence and divergence
∞
X 4
(b) n
n=−1 2
∞
1 3
X
(c) n
+ n+1
n=2 2 2
174
7.5. Changing indices
A useful technique to master is the ability to change the starting index of a se-
ries.
175
7.5. Changing indices
176
7.6. Homework exercises
1. 1 − 31 + 51 − 71 + 19 − ... 7. 1
16
− 1
64
+ 1
44
− 1
45
+ ...
3 4 5 6 7
2. 25
+ 125
+ 625
+ 55
+ 56
+ ... 8. 1 + 1 + 12 + 61 + 1
24
+ 1
120
+ 1
720
+ ...
3. 1 + 31 + 91 + 1
27
+ 1
81
+ 1
35
+ ... 9. 1
14
+ 1
17
+ 1
20
+ 1
23
+ ...
1
4. −2 + 2 − 2 + 2 − 2 + 2 − 2 + 2... 10. 2
+ 24 + 83 + 4
16
+ 5
32
+ ...
4 7 10 13 16 19 1 1 1 1
5. 8
+ 15
+ 22
+ 29
+ 36
+ 43
+ ... 11. 3
− 2·32
+ 3·33
− 4·34
+ ...
6. − 29 − 2
25
− 2
49
− 2
81
− 2
121
− ... 12. 1
2
+ 16 + 1
12
+ 1
20
+ ...
In problems 13-20, you are given an infinite series in Σ−notation. For each series,
find the third term of the series (if it exists), find the ninth term of the series, and
find the fourth partial sum of the series.
∞ ∞
2n−1
13. 17.
P P
n
n!
n=1 n=0
∞ ∞
3
14. [1 + (−1)n ] 18. (−1)n 2n−1
P P
n=0 n=2
∞ ∞
1 1
15. 19.
P P
n n2 +n
n=4 n=1
∞ ∞
cos(πn) (−1)n
16. 20.
P P
n+1 n
n=1 n=1
In problems 21-28, you are given that some infinite series converges to some num-
ber. Then you are given a second infinite series which relates to the first series in
some way. Find the sum of the second series.
∞ ∞
1 1
21. Given = e, compute .
P P
n! n!
n=0 n=2
∞ ∞
1 π2 3
22. Given , compute .
P P
n2
= 6 n2
n=1 n=1
∞ ∞
(−1)n π 2n (−1)n 3π 2n
23. Given = −1, compute .
P P
(2n)! (2n)!
n=0 n=2
∞ ∞
e−2 2n 2n+3
24. Given = 1, compute .
P P
n! n!
n=0 n=3
177
7.6. Homework exercises
∞ ∞
5n 5n
25. Given = e5 , compute .
P P
n! (n+2)!
n=0 n=2
∞ ∞
1 3
26. Given = 2, compute .
P P
2n 2n
n=0 n=−3
∞ n ∞ n+2
1 1
27. Given = 2, compute .
P P
n 2
n 2
n=0 n=2
In problems 29-36, you are given an infinite series. Rewrite each series in Σ−notation
such that the initial term of each rewritten series corresponds to the given index.
∞
4
29. ; starting index 0
P
(n+1)(n+2)
n=3
∞
32n−1
30. ; starting index 4
P
n!
n=2
∞
(−1)n+1
31. ; starting index 2
P
n2n
n=1
∞
2n−1
32. ; starting index 1
P
(n−2)3 −n
n=5
∞
1
33. (−1)n e−3n ; starting index 0
P
n=1
34. 18 − 6 + 2 − 32 + 29 − 2
27
+ ...; starting index 3
178
7.6. Homework exercises
Answers
Note: there are multiple correct answers to numbers 1-12.
∞ ∞ ∞
(−1)n 3n+1 1
1. 5. 9.
P P P
2n+1 7n+1 3n+11
n=0 n=1 n=1
∞ ∞ ∞
n −2 n
2. 6. 10.
P P P
5n−1 (2n+1)2 2n
n=3 n=1 n=1
∞ ∞ n ∞
1 −1 (−1)n+1
3. 7. 11.
P P P
3n 4 n3n
n=0 n=2 n=1
∞ ∞ ∞
1 1
4. 2 · (−1)n 8. 12.
P P P
n! n(n+1)
n=1 n=0 n=1
17. third term is 6; ninth term is 362880; fourth partial sum is 34.
−3 −3 29
18. third term is 5
; ninth term is 17
; fourth partial sum is 35
1 1 4
19. third term is 12
; ninth term is 90
; fourth partial sum is 5
−1 −1 −7
20. third term is 3
; ninth term is 9
; fourth partial sum is 12
π 26 ∞
21. e − 2 28. 2
− 15 33.
P 1
(−1)n+1 e−3n−3
n=0
π2 ∞
22. 4
29.
P
2
(n+4)(n+5) ∞ n
3 2 n=0 1
23. −6 + π 34. (−1)n+1 486
P
2 3
∞ n=3
32n−5
30.
P
24. 8e2 − 40 n=4
(n−2)!
∞
2
1 2 118 35.
P
25. 25
e − 75
∞
(−1)n 2n−7
31.
P
n=4
(n−1)2n−1
n=2
26. 48
∞ ∞
2n+7 n+3
32. 36. (−1)n 3n+8
P P
3
27. 8 n=1
(n+2)3 −n−4
n=0
179
Chapter 8
8.1 Definitions
By far, the most important class of infinite series are geometric series:
Definition 8.1 A series an is called geometric if there exists a real number r such
P
that an+1 = ran for all n. The number r is called the common ratio of the series.
Consequence:
180
8.1. Definitions
If we change notation and call the number a0 just “a”, we get the following im-
portant characterization of geometric series:
n=0
where a is the initial term of the series and r is the common ratio of the series.
In other words, every geometric series is the sum of a constant times all the
nonnegative powers of the common ratio.
If you are given a geometric series, you can always rewrite the series in the
standard form given in the theorem above, where the starting index is n = 0. To
study a geometric series, your first step should almost always be to rewrite the
series in this standard form. To do this, simply write the terms out and factor out
the initial term of the series.
181
8.1. Definitions
Example: For each series, determine if the series is geometric. If it is, write the
series in standard form and identify the common ratio of the series.
∞ n+1
X 7
(a) 3·
n=1 8
4 8 16 32
(b) 2− + − + −...
3 9 27 81
1 1 1
(c) 1+ + + +...
2 3 4
182
8.2. The Geometric Series Test
Theorem 8.3 (Geometric Series Test) Consider a geometric series written in stan-
∞
dard form arn . Then:
P
n=0
Proof of the Geometric Series Test: There are several cases to consider:
Case 1: a = 0
In this case every term of the series is 0 (since they are all of the form arn ).
Therefore the N th partial sum is
N N
arn =
X X
SN = 0=0
n=0 n=0
Case 2: a 6= 0, r = 1
N N N
In this case, SN = arn = a1n =
P P P
a = a + a + ... + a = a(N + 1).
n=0 n=0 n=0
Therefore lim SN = lim a(N + 1) = ±∞ so the series an diverges.
P
N →∞ N →∞
Case 3: a 6= 0, r = −1
183
8.2. The Geometric Series Test
Case 4: a 6= 0, |r| =
6 1
184
8.2. The Geometric Series Test
Remark 1: Since you can always factor a constant a out of a series, the simplest
way to state the content of the Geometric Series Test is as follows:
∞
(
1
X
n = 1−r if |r| < 1
r
n=0
diverges if |r| ≥ 1
(If you learn nothing else from me this semester, learn this fact.)
Remark 2: The geometric series with a = 1 and r = −1 is the series
∞
1(−1)n = 1 − 1 + 1 − 1 + 1 − 1 + 1 − 1 + ...
X
n=0
Theorem 8.4 (Finite Sum Formula for a Geometric Series) Consider a geomet-
∞
ric series written in standard form arn where r 6= 1. Then the N th partial sum
P
n=0
satisfies
N
X
n 2 a(1 − rN +1 )
3 N
ar = a[1 + r + r + r + ... + r ] = .
n=0 1−r
185
8.2. The Geometric Series Test
Examples: For each finite or infinite geometric series, find the sum (if the series
diverges or is not geometric, say so):
2 2 2 2
(a) + + 5 + 7 + ...
3 27 3 3
1 1 1
(b) 4 − 2 + 1 − + − + ...
2 4 8
(c) 2 + 6 + 18 + 54 + ...
186
8.2. The Geometric Series Test
(d) 2+6+18+54+...+2·335
15 n
X 2
(e) 4
n=2 3
1 1 1 1
(f) 1− + − + −...
2 3 4 5
187
8.2. The Geometric Series Test
∞
−2 n
X
(g)
n=0 5
−2
We have a = 1, r = 5
so this series converges to
a 1 1 5
= = = .
1−r 1 − (−2/5) 7/5 7
∞
X 2 · 15n+3
(h) 2n−1
n=1 7
∞
X 7(−1)n 32n−1
(i)
n=2 5 · 24n+3
188
8.3. Applications of geometric series
To answer this question, we first have to start with a side question. Suppose
you invest $P once into an account which pays interest rate R, compounded once
per time period. How much do you have after n time periods?
189
8.3. Applications of geometric series
Repeating decimals
Example: Write the repeating decimal .243737373737... as a fraction in lowest terms.
190
8.3. Applications of geometric series
Pharmacokinetics
Example: Suppose you give a patient a 10 mg dose of a drug daily. If the patient’s
bodily functions remove 90% of whatever amount of that drug is in the patient’s
body,
(a) How much will be in the patient’s body after 14 days (just before he takes the
15th dose)?
(b) How much will end up in the patient’s body (just before he takes each dose)
if he takes the dose indefinitely?
191
8.3. Applications of geometric series
Fractal geometry
Example: Consider the Koch snowflake, a figure constructed by the following
iterative process: first,
√
start with an equilateral triangle of side length 1 (the area of
3
such a triangle is 4 ).
Second, divide each side of the trangle into three segments of equal length and
attach an equilateral triangle to the middle of each segment. Then erase the middle
of each segment. After doing this, you get the following figure:
Third, repeat this procedure indefinitely. This means that at each stage, you
take each side of the figure, divide it into thirds, attach an equilateral triangle to the
middle third of each segment, and the erase the middle of each previous segment.
If you carry out this procedure, you get the following sequence of figures in the
next three steps:
Repeating this procedure infinitely many times produces a figure called the
Koch snowflake. The question is to find the perimeter of the Koch snowflake, and
find the area enclosed by the Koch snowflake.
192
8.3. Applications of geometric series
SIDE
# OF LENGTH AREA
SIDES OF OF TOTAL ADDITIONAL
#∆S IN EACH ∆ EACH ∆ AREA PERIMETER
STAGE ADDED FIGURE ADDED ADDED ADDED ADDED
√ √
3 3
0 1 3 1 4 4
3
..
.
193
8.4. The Ratio Test
In this section we start trying to solve the first of these questions - determining
whether or not a series converges.
In this packet we develop our first test, based on reasoning coming from geo-
metric series.
a1 a2 a3 a4
= = = = ... = r.
a0 a1 a2 a3
∞
All geometric series can be rewritten as arn ; by the GST we have
P
n=0
∞
(
a
X
n = 1−r if |r| < 1
ar .
n=0
diverges if |r| ≥ 1
Now let an be any (not necessarily geometric) infinite series. In this setting
P
a1 a2 a3 a4
, , , , etc.
a0 a1 a2 a3
are not likely to be the same number. However, it may be the case that these ratios
get closer and closer to some number ρ (this is the Greek letter “rho”), i.e.
an+1
lim
n→∞ a
= ρ.
n
194
8.4. The Ratio Test
This means that for large n, an+1 ≈ ρ so an should behave like a geometric series
P
an
with common ratio very, very close to ρ.
Consequences:
1. If ρ > 1, then an behaves like a geometric series with |r| > 1, i.e.
P
2. If ρ < 1, then an behaves like a geometric series with |r| < 1, i.e.
P
The conclusion of all this logic is what is called the Ratio Test:
|an+1 |
Theorem 8.5 (Ratio Test) Suppose an is an infinite series and let ρ = lim .
P
n→∞ |an |
Then:
Remarks:
2. Since you are taking the limit of ratios which are positive, the value of ρ must
be nonnnegative. If you get ρ < 0, you have done something wrong (you
forgot the absolute values inside the limit).
195
8.4. The Ratio Test
Types of series for which the Ratio Test will not work well:
196
8.4. The Ratio Test
197
8.4. The Ratio Test
198
8.4. The Ratio Test
199
8.4. The Ratio Test
Remark: Notice that the evaluation of ρ in the Ratio Test often contains either a
simplification of exponents of the form
cn+1 cn 1
=c or = etc.
cn cn+1 c
or a simplification of factorials of the form
n! 1 (n + 1)! (2n)! 1
= or =n or = etc.
(n + 1)! n+1 n! (2n + 2)! (2n + 2)(2n + 1)
When simplifying factorial expressions, it is often useful to write out the terms
being multiplied to see how they will be cancelled. For example, the last equality
above comes from
(2n)! 2n(2n − 1)(2n − 2) · · · 3 · 2 · 1 1
= = .
(2n + 2)! (2n + 2)(2n + 1)2n(2n − 1)(2n − 2) · · · 3 · 2 · 1 (2n + 2)(2n + 1)
In general,
200
8.5. Homework exercises
∞
n+2
∞
−3 5 3
10. −
P
2.
P
6n (−1)n 3n 5
n=0 n=1
3. 9 − 92 + 49 − 89 + 9
− 9
+ ... 11. 80 + 40 + 20 + 10 + 5 + 52 + 54 + ...
16 32
∞
17 3·8n
12.
P
2
4.
P
52n−3
5n n=0
n=3
∞
2·32n−1
13.
P
11 n
3 5·24n+3
5.
P
n=2
42n
n=2
∞
14. 3−n
P
6. 2 + 4 + 8 + 16 + ... + 2100 n=0
∞ n ∞
−4 3n −5
7.
P
15.
P
3 6n
n=3 n=1
∞ ∞
6n−1
8. 16. 4n 5−n
P P
7n+1
n=2 n=0
17. Suppose the initial term of a geometric series is 3 and the series converges to
2. What is the common ratio of the series?
2
18. A geometric series with common ratio 3
converges to 8. What is the initial
term of the series?
23. A ball is dropped from a height of 15 ft onto a concrete slab. Each time the
ball bounces, it rebounds directly to 23 of its previous height. Find the total
distance the ball travels before it comes to rest.
24. Suppose a patient takes 25mg of a certain drug each day. If 80% of the drug
is excreted by bodily functions each day, how much of the drug will be in the
patient’s body immediately before she takes her 22nd dose of the medicine?
201
8.5. Homework exercises
25. Suppose your drinking water contains poison, and as such you injest 2.3mg
of the poison each day. Although your body gets rid of 10% of the poison
in your body each day, when you accumulate 20mg of the poison in your
system you will be dead. How long do you have before you need to stop
drinking your water?
26. Suppose your drinking water contains poison, and as such you injest 2.3mg
of the poison each day. Suppose further that when you accumulate 250mg of
the poison in your system you will be dead. What percent of the poison in
your system does your body need to excrete daily in order to never die from
the poison? (Assume that you will live forever if the poison doesn’t kill you.)
27. In the figure below, the triangle indicated by the solid lines is an isoceles
right triangle whose height is 1 unit. All the dashed line segments drawn
are perpendicular to either the base of the triangle or the hypotenuse. If the
dashed line segments continue indefinitely, find the total length of the dashed
line segments.
202
8.5. Homework exercises
Answers
1 1313
1. 4
20. 999
−18 634
2. 5 21. 2475
3. 6 22. 1319
9990
15
1 1
4. 50
1− 5 23. 75 ft
24. 6.25 mg
10
9 3
5. 208
1− 16
25. On the 19th day you will die.
6. 2(2100 − 1)
26. .92%
7. diverges √
6
27. 1 + 2
8. 49
28. diverges
9. 9
−179 29. diverges
10. 100
30. converges
11. 160
9375 31. diverges
12. 17
203
Chapter 9
This chapter deals only with question 1 above: determining whether or not a
series converges.
We know two tests so far which tell us whether some series converge or di-
verge:
1.
2.
204
9.1. Classifying series according to sign
What do you do next, if these tests don’t apply? Next, look at the signs of the
individual terms of the series. This is because whether or not any of the rest of our
tests are valid depends in part on what the signs of the series are.
Definition 9.1 An infinite series is called positive if all its terms are positive; more
precisely an is positive if an ≥ 0 for all n.
P
Definition 9.2 An infinite series is called negative if all its terms are negative; more
precisely an is negative if an ≤ 0 for all n.
P
205
9.2. The Integral Test; harmonic and p-series
2. f is decreasing.
Then:
R∞ ∞
• if f (x) dx converges, then an converges, and
P
0
n=0
R∞ ∞
• if f (x) dx diverges, then an diverges.
P
0
n=0
0 1 2 3 4 5 6 7 8 9 10
206
9.2. The Integral Test; harmonic and p-series
2. The easiest way to check that a function f is decreasing is to show that the
derivative f 0 is negative.
3. If your series contains terms like n!, the Integral Test cannot be used (because
you cannot define a function with an x! in it whose domain contains non-
integers).
4. Although the statement of the Integral Test above uses the starting index 0 for
both the integral and the series, this is really not important (since the starting
index of a series is irrelevant).
6. The Integral Test should be used only as a last resort (other tests are better).
207
9.2. The Integral Test; harmonic and p-series
Recall that when studying improper integrals we came across the following
fact: Z ∞ (
1 converges if p > 1
dx
1 x p diverges if p ≤ 1
Putting this fact together with the Integral Test, we have:
1. an converges if p > 1.
P
2. an diverges if p ≤ 1.
P
∞
7
Example:
P
n6
n=5
∞
−8
Example:
P
√
n
n=2
∞
Example: 1 + 21 + 13 + ... = 1
P
n
n=1
208
9.2. The Integral Test; harmonic and p-series
Proof:
Z ∞ b
A A A A
= lim ln(Bx + C) = lim ln(Bb + C) − ln(B + C) = ±∞
1 Bx + C b→∞ B 1 b→∞ B B
A
so by the Integral Test, the series diverges.
P
Bn+C
Note:
∞
2
(b)
P
3n+4
n=3
∞
2 3
(c) −
P
n n3
n=3
∞
−2
(d)
P
√
n
n=3
209
9.3. The nth -term Test
∞
2n2
(e)
P
5n7
n=3
then an diverges.
P
Important:
Theorem 9.10 (nth Term Test, restated) Suppose an is an infinite series. If lim
P
n→∞
|an | =
6 0 (this includes when this limit is ∞ or DNE), then an diverges.
P
210
9.3. The nth -term Test
∞
X 1
(a)
n=2 n2 +1
∞
3 7
X
(b) n
− 2
n=1 2 n
∞
X n+1
(c)
n=0 n + 3
211
9.4. The Comparison Test
Remarks:
Idea: Use this test with series that are similar to a “simpler” series (usually the
simpler series is a geometric or p−series). Reason as follows:
212
9.4. The Comparison Test
3. Series whose terms contain sines or cosines. If the terms contain some ex-
pression of the form cos or sin
∞
X 5
(a)
n=1 n3 + 3n + 8
213
9.4. The Comparison Test
∞
1 1
X
(b) −
n=1 n n+4
∞
X 3 − cos n
(c)
n=1 n
∞
3n + 7n2
X
(d)
n=1 12n6
∞
X n
(e)
n=0 n4 +1
214
9.4. The Comparison Test
∞
X 2
(f) √
n=10
3
n−3
∞
2
e−n
X
(g)
n=1
∞
X (−1)n+1
(h)
n=1 n
215
9.5. Homework exercises
∞ ∞
3 1 1
2. 13. −
P P
5n−3 k k+1
n=1 k=1
∞ ∞
−3 1
3. 14.
P P
√
2n n ln(2n )
n=1 n=4
∞ ∞
2+cos(3n)
4. e−1/n 15.
P P
n
n=2 n=1
∞ ∞
2n2 +3 1
16.
P
5.
P
5n2 −4 en +e−n
n=2 n=0
∞ ∞
4 2 3+cos(2n)
17.
P
6.
P
n5
+ n 4n2
n=1 n=0
∞ ∞
ln n 3+cos(2n)
18.
P
7.
P
√
n 43n
n=3 n=0
∞ ∞
3+2n
19.
P
8.
P
ln n 3n +4
n=1 n=2
∞ ∞
arctan n 4+sin(n2 +2n)
20.
P
9.
P
√ 3
n2 +1 n=1 n5 +1
n=1
∞ ∞ 2 −3n
(n − 1)−1/2 21. 5−n
P
10.
P
n=2 n=2
∞ ∞
2 6n
11. 3ne−n 23.
P P
62n +3
n=1 n=1
216
9.5. Homework exercises
Answers
217
Chapter 10
diverges.
This series is called the alternating harmonic series. Does it converge or diverge?
• lim |an | =
n→∞
218
10.1. Alternating series
Now what?
∞
(−1)n+1
Let’s compute the partial sums of the alternating harmonic series :
P
n
n=1
S 1 = a1 = 1
1
S 2 = a1 + a2 = 1 − =
2
1 1
S 3 = a1 + a2 + a3 = 1 − + =
2 3
S4 =
Sn
S1 =1
S3 =56
S5
S6
S4
S2 =12
n
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Observe:
219
10.1. Alternating series
such that:
1. an is alternating (i.e. the terms being added alternate between positive and
P
negative);
Then an converges.
P
Important:
lim |an |.
3. Otherwise, compute n→∞
• If this limit is nonzero, the series diverges by the nth -term Test.
• If this limit is zero, verify that |an | ≥ |an+1 | (it will in Math 230). Then
the series converges by the AST.
220
10.1. Alternating series
n=2 n+4
221
10.2. The triangle inequality
Problem: This series is not geometric, not a p−series, not positive (so the In-
tegral and Comparison Tests are no good) and not alternating (so the Alternating
Series Test is no good). The only test we know so far that we could use to study
this is the nth −term Test:
A new idea in the study of series is the following important concept about num-
bers:
Theorem 10.2 (Triangle Inequality for R) For all real numbers a and b, |a + b| ≤
|a| + |b|.
ab ≤ |ab|
2ab ≤ 2|ab|
a + 2ab + b2 ≤ |a|2 + 2|ab| + |b|2
2
222
10.2. The triangle inequality
Theorem 10.3 (Generalized Triangle Inequality for R) For any finite list of real
numbers a1 , a2 , ..., an ,
n n
X X
aj ≤ |aj |.
j=1 j=1
X sin(en ) X 1 n
−n −n −n
X X X
|an | = | sin(en )|
e = |±1| e = e =
e
Proof of the Triangle Inequality for Infinite Series: Let bn = an + |an |. Notice
that if an < 0, an +|an | = an −an = 0 and if an ≥ 0, then an +|an | = |an |+|an | = 2|an |.
Thus bn is always either 0 or 2|an |. In particular, we have
0 ≤ bn ≤ 2|an |.
see X X X X
an = (bn − |an |) = bn − |an |
is the difference of two convergent series, hence an converges.
P
223
10.3. Absolute and conditional convergence
Remarks:
• By the Triangle Inequality for Infinite Series, we know that if a series is abso-
lutely convergent, then it converges.
absolutely converges.
• Based on these observations, there are three possibilities for an infinite series
an :
P
3. Something else (which given the remarks above must be that an con-
P
then we say an is conditionally convergent (or that the series converges con-
P
ditionally).
Based on the ideas developed thus far, we can now create a Venn diagram
which incorporates the sign of a series together with whether the series converges
absolutely, converges conditionally, or diverges. This Venn diagram is on the next
page:
224
10.3. Absolute and conditional convergence
CONVERGENT SERIES
POSITIVE ALTERNATING
SERIES SERIES
COND’L.
CONV’T
ABSOLUTELY
CONVERGENT
NEITHER
POSITIVE NOR
ALTERNATING
ALL INFINITE SERIES
We can now refine one of our “big picture” questions with infinite series:
For now, we turn our attention to the third question: the rearrangement prob-
lem.
225
10.4. Rearrangement of infinite series
Since the series alternates and has positive initial term, L ∈ [ 12 , 1].
226
10.4. Rearrangement of infinite series
Further investigation: Look only at the terms of the alternating harmonic series
which are positive:
∞
1 1 1 X 1
1 + + + + ... =
3 5 7 n=1 2n + 1
Now look only at the terms of the alternating harmonic series which are negative:
∞ ∞
1 1 1 1 1 X −1 −1 X 1
− − − − − − ... = =
2 4 6 8 10 n=1 2n 2 n=1 n
Consequence:
Loosely speaking, the problem is that this series converges conditionally (as
opposed to absolutely). Any conditionally convergent series is comprised of an
infinite amount of “positive stuff” and an infinite amount of “negative stuff”, and
the series only converges because the terms cancel each other out in a very delicate
way. On the other hand, an absolutely convergent series (almost by definition) has
only a finite amount of “positive stuff” and a finite amount of “negative stuff”, so
no matter how you rearrange the series you always get the same thing. To sum-
marize:
so that the rearranged series converges to any number you like! (The series can
also be rearranged so that the rearranged series diverges.)
2. If an converges absolutely to L, then no matter how the terms of the series are
P
227
10.4. Rearrangement of infinite series
Additional remarks:
2. If a series is shown to converge by the Ratio Test, then the series converges
absolutely (since when evaluating ρ one takes absolute values of the terms in
the series).
In particular, for any conditionally convergent series, if you tried the Ratio
Test you would always get ρ = 1. (However, if ρ = 1 the series could con-
verge absolutely, converge conditionally or diverge.)
3. p-series converge absolutely when they converge (since they are positive).
5. Since the Integral Test and Comparison Test apply only to positive series, if
you use one of these to show that a series converges, then the series must
converge absolutely.
6. The Alternating Series Test tells you only that a series converges (not whether
the series converges absolutely or conditionally). To distinguish between
these situations, you need further analysis.
Putting together everything we know, we have the outline on the next page
which describes how to determine whether a given series is absolutely convergent,
conditionally convergent or divergent.
228
10.5. Summary of classification techniques
229
10.6. Examples
10.6 Examples
In each of the following examples, you are to classify the given series as ab-
solutely convergent, conditionally convergent, or divergent, giving appropriate
justification.
∞
X (−1)n nn
(a)
n=1 4n
230
10.6. Examples
∞
(−1)n n−1/3
X
(b)
n=1
∞
X 2n4
(c) 8 11
n=1 n + 3n + 12
231
10.6. Examples
∞
(−1)n
!
X 3
(d) 2
+ n
n=1 n 4
1 1 1 1 1
(e) 1− + − + − 5 +...
5 25 125 625 5
232
10.7. Homework exercises
∞ ∞
(−1)n n2 4 cos(πn)
2. 5.
P P
3n 2n +3
n=2 n=3
∞ ∞
4
5n +3 (−1)n n2
3. (−1)n 20n4 +2n 6.
P P
2 +n+1 ln(n+4)
n=4 n=0
21. If n→∞
lim an = 0, then an converges.
P
233
10.7. Homework exercises
∞
n+1
24.
P
ln(2n−5)
n=4
∞
2 cos(πk)
25.
P
k2
k=1
∞ n
(−1)
26.
P
√
2 n+2
n=0
∞
27. (−1)n+1 n2n+1
P
n=1
∞
(−1)n 2n
28.
P
n3 +5
n=0
∞
(−1)n 42n
29.
P
7n n!
n=0
∞
30. 4(−1)n+1 n−3/5
P
n=1
∞ h
(−1)n
i
3
31.
P
n5
+ n
n=0
∞ 1
(−1)[ 2 n(n+1)]
32.
P
4n +n2
n=1
∞
n
33. (Challenge)
P
ln n+1
n=1
1 1 1 1
34. (Challenge) 1 + 1.1
+ 1.11
+ 1.111
+ 1.1111
+ ...
234
10.7. Homework exercises
Answers
235
Chapter 11
Taylor series
f (x) = a0 + a1 x + a2 x2 + a3 x3 + a4 x4 + a5 x5 + ...
f 0 (x) = a1 + 2a2 x + 3a3 x2 + 4a4 x3 + 5a5 x4 + ...
f 00 (x) = 2a2 + 3 · 2a3 x + 4 · 3a4 x2 + 5 · 4a5 x3 + ...
f 000 (x) = 3 · 2a3 + 4 · 3 · 2a4 x + 5 · 4 · 3a5 x2 + ...
.. .
. = ..
f (n) (x) = n · (n − 1) · (n − 2) · · · 3 · 2 · 1an + (const)x + (const)x2 + ...
236
11.1. Uniqueness of power series
f (0) =
f 0 (0) =
f 00 (0) =
f 000 (0) =
.. ..
. = .
f (n) (0) =
f (n) (0)
an = for all n.
n!
In other words, the only power series of the form an xn which can represent f is the
P
series
∞
X f (n) (0) n
x .
n=0 n!
This series is called the Taylor series (centered at 0) of f or the Maclaurin series
of f .
Corollary 11.2 Suppose you have two power series which represent the same function
of x, i.e.
∞ ∞
an x n = bn x n
X X
n=0 n=0
237
11.1. Uniqueness of power series
Note: There are functions which cannot be represented as a power series. How-
ever, we will not encounter those in Math 230 (so we will assume that every func-
tion f is equal to its Taylor series at all x for which the Taylor series converges).
Example 1: f (x) = ex .
Note that since the exponential function is its own derivative, we have f (n) (x) =
ex for all n and we know therefore that f (n) (0) = e0 = 1 for all n. Thus, for all n, we
have
an =
One can show that this series converges for all x using the Ratio Test.
238
11.1. Uniqueness of power series
∞
xn x2 x3 x4
ex =
P
n!
=1+x+ 2
+ 3!
+ 4!
+ ...
n=0
239
11.1. Uniqueness of power series
This series can also be shown to converge for all x using the Ratio Test. We have
obtained the following formula (which must be memorized):
∞
(−1)n x2n+1 x3 x5 x7 x9
=x− − − ...
P
sin x = (2n+1)! 3!
+ 5! 7!
+ 9!
n=0
We could go through the procedure of Example 2 again and find a pattern with
the an . But there is a better way:
∞
(−1)n x2n x2 x4 x6 x8
=1− − − ...
P
cos x = (2n)! 2
+ 4! 6!
+ 8!
n=0
240
11.1. Uniqueness of power series
There is one other function whose Taylor series we “know” (even though we
didn’t call it a Taylor series at the time). Suppose we set an = 1 for all n. Then we
get
∞ ∞ ∞
an x n = 1xn = xn
X X X
∞
1
xn = 1 + x + x2 + x3 + ... for −1 < x < 1
P
1−x
=
n=0
This last series can be manipulated to find Taylor series of some other functions.
If you replace the x on both sides of the above equation with −x, we get
1
= 1 + x + x2 + x3 + ...
1−x
1
=
1+x
We have derived:
∞
(−1)n+1 n
= x − 12 x2 + 13 x3 − 14 x4 + ... for −1 < x ≤ 1
P
ln(1 + x) = x
n
n=1
1
Now go back to the series for 1−x , and let’s do some different manipulations.
2
First, replace x with −x , and then integrate:
241
11.1. Uniqueness of power series
1
= 1 + x + x2 + x3 + ...
1−x
1
=
1 + x2
We have derived:
∞
(−1)n 2n+1
= x − 31 x3 + 15 x5 − 17 x7 + ... for −1 ≤ x ≤ 1
P
arctan x = 2n+1
x
n=0
242
11.1. Uniqueness of power series
∞
xn x2 x3
ex = + ... (holds for all x)
P
n!
=1+x+ 2
+ 3!
n=0
∞
(−1)n x2n+1 x3 x5 x7
=x− − + ... (holds for all x)
P
sin x = (2n+1)! 3!
+ 5! 7!
n=0
∞
(−1)n x2n x2 x4 x6
=1− − (holds for all x)
P
cos x = (2n)! 2
+ 4! 6!
+ ...
n=0
∞
1
xn = 1 + x + x2 + x3 + ... (holds when x ∈ (−1, 1))
P
1−x
=
n=0
∞
(−1)n+1 n x2 x3 x4
=x− − (holds when x ∈ (−1, 1])
P
ln(1 + x) = x n 2
+ 3 4
+ ...
n=1
∞
(−1)n 2n+1 x3 x5 x7
=x− − + ... (holds when x ∈ [−1, 1])
P
arctan x = x
2n+1 3
+ 5 7
n=0
From these six Taylor series, one can obtain the Taylor series for lots of other
functions via manipulations:
243
11.1. Uniqueness of power series
Example: Find a power series representation (i.e. find the Taylor series) of
each of the following functions (give the answer in Σ−notation, and in written-out
form):
5
(b) f (x) = x2 e−3x
x
(c) f (x) = (1−x)2
4
(f) f (x) = 3+5x
244
11.2. Applications of Taylor series
6 sin x − 6x + x3
lim
x→0 2x5
Old solution using L’Hôpital’s Rule:
6 sin x − 6x + x3 0
lim 5
=
x→0 2x 0
2
L 6 cos x − 6 + 3x 0
= lim 4
=
x→0 10x 0
L −6 sin x + 6x 0
= lim =
x→0 40x3 0
L −6 cos x + 6 0
= lim =
x→0 120x2 0
L 6 sin x 0
= lim =
x→0 240x 0
L 6 cos x 6 1
= lim = = .
x→0 240 240 40
New solution using Taylor series:
245
11.2. Applications of Taylor series
Definition 11.3 Suppose f is a function which can be differentiated over and over
again at x = 0; the Taylor series of f is
∞
X f (n) (0) n
x .
n=0 n!
If we truncate this series at the N th power term, we obtain a partial sum of the Taylor
series called the N th Taylor polynomial (centered at 0) (a.k.a. Taylor polynomial
of order N ) of f . This polynomial is denoted PN (x).
246
11.2. Applications of Taylor series
P0 (x) =
P1 (x) =
P2 (x) =
P3 (x) =
..
.
PN (x) =
P0 (x) =
P1 (x) =
P2 (x) =
P3 (x) =
P4 (x) =
P5 (x) =
247
11.2. Applications of Taylor series
The point of Taylor series and Taylor polynomials is that if a function f has
th
N Taylor polynomial PN , then PN (x) ≈ f (x) (this approximation improves as N
increases, but is pretty good even for small N ). For example, let’s look at the Taylor
polynomials for ex and sin x graphically:
8
-4 -2 2 4
-10 -5 5 10
-1
-2
-3
248
11.2. Applications of Taylor series
Example: Approximate ln(1.2) using the fourth Taylor polynomial (a.k.a. Tay-
lor polynomial of order 4) for an appropriately chosen function.
R1
Example: Approximate 0 sin(x4 ) dx by replacing the integrand with its tenth
Taylor polynomial.
249
11.2. Applications of Taylor series
Higher-order derivatives
Example: Let f (x) = x20 cos(3x8 ). Find f (100) (0), the 100th derivative of f at zero.
250
11.2. Applications of Taylor series
2π 3 2π 5 2π 7
π− + − + ...
23 · 3! 25 · 5! 27 · 7!
251
11.3. General theory of power series
The partial sums of this series are denoted Pn (x) and called the Taylor polynomials
of f . Ideally, f (x) ≈ Pn (x), so f can be replaced by Pn in approximation problems.
n=0
where the an are constants (which may or may not have anything to do with some
function f ). Such functions are called power series (centered at 0). In fact, we can
study more general objects:
where the a0 , a1 , ... are real numbers. A power series (in x) is a power series centered
at x = 0, i.e. a power series is a function of the form
∞
an x n
X
f (x) =
n=0
= a0 + a1 x + a2 x2 + a3 x3 + a4 x4 + ...
The numbers a0 , a1 , a2 , ... are called the coefficients of the power series; the terms of
the series are a0 , a1 x, a2 x2 , ...
252
11.3. General theory of power series
n=0
253
11.3. General theory of power series
254
11.3. General theory of power series
∞
Theorem 11.5 (Cauchy-Hadamard (C-H) Theorem) Let an (x−a)n be a power
P
n=0
series centered at a. There exists a quantity R (which is either a nonnegative real num-
ber or ∞) called the radius of convergence of the series such that:
If you don’t like using the Ratio Test, you can find the value of R by the follow-
ing formula:
∞
Theorem 11.6 (Abel’s Formula) Let an (x − a)n be a power series centered at a.
P
n=0
Then, the radius of convergence of this power series is given by
|an |
R = lim
n→∞ |an+1 |
Remarks:
1. Abel’s Formula uses only the coefficients of the power series and not the powers
of (x − a).
2. Abel’s Formula looks like the computation done in the Ratio Test, but the
fraction used to compute R is “upside-down” compared to what was used to
compute ρ in the Ratio Test.
255
11.3. General theory of power series
According to the C-H Theorem, the set of x for which a power series converges
is an interval; this interval is called the interval of convergence of that series and is
the domain of that series.
|an |
R = lim .
n→∞ |a
n+1 |
3. If R = 0, then the series converges absolutely for x = a and diverges for all
other x. The interval of convergence is [a, a] or just {a}.
Example: For each power series, find the radius of convergence and the set of
x for which the series converges.
∞
X (−1)n (x + 1)n
(a)
n=0 (2n)!
256
11.3. General theory of power series
∞
3n (x − 2)n
X
(b)
n=0
∞
n n xn
X
(c)
n=2
257
11.4. Homework exercises
7. f (x) = −3
−2−x
18. f (x) = (x2 + 2) cos x
Hint: Find the Taylor series of x2 cos x
8. f (x) = e−x
and 2 cos x independently, and then
9. f (x) = cos 4x add them term-by-term.
1
12. f (x) = cos x2 21. f (x) = (1−x)2
For problems 22-32, evaluate the following limits without using L’Hopital’s Rule:
sin x arctan 6x2 −6x2
22. lim 28. lim x6
x→0 x x→0
cos x−1
23. lim x sin x8 −x8
x→0 29. lim x20
x→0
sin x3 −x3
24. lim x9
x→0 sin x8 −x8
30. lim x24
arctan x−x x→0
25. lim x3
x→0
sin x8 −x8
ex −1−x 31. lim x30
26. lim 2x2
x→0
x→0
258
11.4. Homework exercises
In problems 33-37, approximate each of the following numbers using the second
Taylor polynomial of an appropriately chosen function:
In problems 38-42, approximate each of the following numbers using the fourth
Taylor polynomial of an appropriately chosen function:
√
38. sin 14 40. e
46. Approximate 01 ln(2x2 + 1) dx by replacing the integrand with its fourth Tay-
R
lor polynomial.
R1 2
47. Approximate x sin(x2 ) dx
0 by replacing the integrand with its sixth Taylor
polynomial.
48. Approximate 01 x8 sin x dx by replacing the integrand with its twelfth Taylor
R
polynomial. Describe the integration technique that one would use to find
the exact value of this integral. (Isn’t using a Taylor polynomial better?)
259
11.4. Homework exercises
3
52. Find f (40) (0) if f (x) = e2x .
3
53. Find f (42) (0) if f (x) = e2x .
∞
xn
54. Find f (30) (0) if f (x) = .
P
(4n)!
n=0
∞
x4n+2
55. Find f (30) (0) if f (x) = .
P
(4n)!
n=0
∞
58. Determine whether the series cos( n1 ) − 1 converges or diverges.
P
n=1
∞
1
59. Determine whether the series ln 1 − converges or diverges.
P
n
n=1
In problems 60-69, find the sum of each of the following series (you may assume
without proof that the series converge):
60. 1 − 13 + 51 − 71 + 19 − 1
11
+ ...
2 2 2 2
61. 2 + 2 + 2!
+ 3!
+ 4!
+ 5!
+ ...
π π2 π4 π6
62. 4
− 42 2!
+ 44 4!
− 46 6!
+ ...
9 27 81 35
63. 1 − 3 + 2!
− 3!
+ 4!
− 5!
+ ...
e2 e3 e4
64. 1 + e + 2
+ 3!
+ 4!
+ ...
2π 2 2π 4 2π 6
65. 2 − 2!
+ 4!
− 6!
+ ...
1 1 1 1
66. 1 − 2!
+ 4!
− 6!
+ 8!
− ...
100 10000 106 108
67. 2!
− 4!
+ 6!
− 8!
+ ...
π2 π4 π6 π8
68. 1 − 22 3!
+ 24 5!
− 26 7!
+ 28 9!
− ...
1 1 1 1
69. 2
− 22 ·2
+ 23 ·3
− 24 ·4
+ ...
260
11.4. Homework exercises
Answers
∞
1. 3xn = 3 + 3x + 3x2 + 3x3 + 3x4 + ...
P
n=0
∞
2. x2n = 1 + x2 + x4 + x6 + ...
P
n=0
∞
(−1)n+1 n x2 x3 x4 x5
3. =x− − − ...
P
x n 2
+ 3 4
+ 5
n=1
∞
4. (n + 1)(n + 2)xn = 2 + 6x + 12x2 + 20x3 + 30x4 + ...
P
n=0
∞
5. x3n = 1 + x3 + x6 + x9 + ...
P
n=0
∞ n 2 3 4
−5
6. xn = 1 − 52 x + 5
x2 − 5
x3 5
x4 − ...
P
2 2 2
+ 2
n=0
∞ n 2 3 4
3 −1 3
7. xn − 32 · 21 x + 3 1
x2 − 3 1
x3 + 3 1
x4 − ...
P
2 2
= 2 2 2 2 2 2 2
n=0
∞
(−1)n n x2 x3 x4
8. =1−x+ − − ...
P
x n! 2 3!
+ 4!
n=0
∞
(−1)n 42n 2n 42 2 44 4 46 6
9. =1− −
P
x
(2n)! 2!
x + 4!
x 6!
x + ...
n=0
∞
−2n n
10. = −2x − 42 x2 − 83 x3 − 16 4
− 32 5
− ...
P
x n 4
x 5
x
n=1
∞
11. (−2)n xn = 1 − 2x + 4x2 − 8x3 + 16x4 − ...
P
n=0
∞
(−1)n 4n x4 x8 x12
12. =1− −
P
x
(2n)! 2!
+ 4! 6!
+ ...
n=0
∞
−1 2n
13. = −x2 − 21 x4 − 13 x6 − 14 x8 15 x10 − 16 x12 + ...
P
x n
n=1
∞
(−1)n 32n+1 2n+2 3 2 33 4 35 6 37 8 39 10
14. − − − ...
P
x
(2n+1)!
= 1!
x 3!
x + 5!
x 7!
x + 9!
x
n=0
∞
(−1)n 2n+1 x5 x7 x9
15. = x − x3 + − − ...
P
x n! 2! 3!
+ 4!
n=0
∞
(−1)n 4n+3
16. = − 3!1 x7 + 5!1 x11 − 7!1 x15 + 9!1 x19 − ...
P
(2n+1)!
x
n=1
∞
2
17. x2n = 2 + x2 + 4!2 x4 + 6!2 x6 + ...
P
(2n)!
n=0
261
11.4. Homework exercises
∞
1 2 2 1 2
18. 2+ (−1)n+1 − x2n+2 = 2 + 1 − x2 − − x4 + ...
P
(2n)! (2n+2)! 2! 2! 4!
n=0
∞
(−1)n 4n+3 x7 x11 x15
19. = x3 − −
P
x
2n+1 3
+ 5 7
+ ...
n=0
∞
20. 2(−1)n x2n = 2 − 2x2 + 2x4 − 2x6 + 2x8 − ...
P
n=0
∞
21. (n + 1)xn = 1 + 2x + 3x2 + 4x3 + 5x4 + ...
P
n=0
14701 30!
22. 1 39. 15000
54. 120!
211
23. 0 40. 128
55. 870
−1 11
24. 6 41. 24 56. f (0) = 2; f 0 (0) = −1;
−1 179
f 00 (0) = −2
3
; f 000 (0) = 12.
25. 3 42. 128
1 57. diverges
26. 4 43. 9
20
1 58. converges
27. 3 44. 1
24
59. diverges
28. −72 45. 2
π
29. 0 4
60. 4
46. 15
30. −1 1
61. 2e
6 47. 5 √
2
31. −∞ 2557 62. 2
48. 14040
;
32. 1
use parts 8 times to 63. e−3
2
262
Index
PN (x), 246 center of mass, idea of, 100
nth Term Test, 210 center of mass, one-dimensional con-
p-Series Test, 208 tinuous, 104
p-series (definition), 208 center of mass, one-dimensional dis-
“big six” Taylor series, 243 crete, 103
center of mass, two-dimensional con-
Abel’s Formula, 255 tinuous, 107, 108
absolutely convergent, 224 center of mass, two-dimensional dis-
acceleration (parametric equations), 145 crete, 106
alternating series, 220 centroid, 110
Alternating Series Test, 220 Chain Rule, 12
antiderivative, 15 circle, 134
applications of integration, general prin- circle, parametric equations of, 134
ciple, 96 classification problem (series), 164, 168,
arc length, 97 194, 204
Arc length formula (Cartesian equa- classifying a series, methods for, 229
tion), 98 coefficients (of a power series), 252
Arc length formula (parametric equa- common ratio, 180
tions), 147 Comparison Test (series), 212
area, 80 Comparison Test for Improper Integrals,
associative property, 165 73
AST, 220 complicated u-substitutions, 35
bounded (region), 61 computation problem (series), 164, 168,
194, 204
C-H Theorem, 255 concavity, 14
calculus formulas for parametric equa- Concavity (of parametric equations),
tions, 152 151
Cartesian function, parametric equa- conditionally convergent, 224
tions of, 141 Constant Multiple Rule, 12
Cauchy-Hadamard Theorem, 255 continuous (function), 8
263
Index
264
Index
265
Index
tangent line, 14
Taylor polynomial of order N , 246
Taylor series, definition, 237
Taylor series, list of, 243
terms (of a series), 252
tone, 14
total mass, one-dimensional continu-
ous, 104
total mass, one-dimensional discrete,
103
total mass, two-dimensional continu-
ous, 107
total mass, two-dimensional continu-
ous, 108
total mass, two-dimensional discrete,
106
266