Calculus 3 Module 2
Calculus 3 Module 2
Calculus 3 Module 2
EDUCATION
COURSE MODULE IN
CALCULUS 3
(EDMM113)
1st Semester, A.Y. 2021 – 2022
2
COURSE FACILITATOR: PONCIANO A. GALE, JR.,
LPT, MEd-MATH
FB/MESSENGER: Ponciano A. Gale Jr.
Email: poncianogalejr81@gmail.com
Phone No: 09659700806
MODUL
This document is a property of NONESCOST Module 1 | Page 1
Unauthorized copying and / or editing is prohibited. (For Classroom Use Only) Prepared by: Ponciano A.Gale, Jr.
VISION
MISSION
research and resource creation opportunities in various technical and disciplinal areas.
GOAL
INSTITUTIONAL OUTCOMES
community.
Warm greetings!
Welcome to the 1st semester of School Year 2021-2022! Welcome to the College of
Education and welcome to NONESCOST!
Despite of all the happenings around us, there is still so much to be thankful for and one
of these is the opportunity to continue learning.
You are right now browsing your course module in EDMM113. As you read on, you will
have an overview of the course, the content, requirements and other related information
regarding the course. The course is made up of 6 lessons. Each lesson has seven
parts:
LEARNING ACTIVITIES – To measure your learnings in the lesson where you wandered
Please read your modules and learn the concepts by heart. It would help you prepare to
be effective and efficient professional in your respective fields. You can explore more of
the concepts by reading the references and the supplementary readings.
I encourage you to get in touch with me in case you may encounter problems while
studying your modules. Keep a constant and open communication. Use your real
names in your FB accounts or messenger so I can recognize you based on the list of
officially enrolled students in the course. I would be very glad to assist you in your
journey. Furthermore, I would also suggest that you build a workgroup among your
classmates. Participate actively in our discussion board or online discussion if possible
and submit your outputs/requirements on time. You may submit them online through
email and messenger. You can also submit hard copies. Place them in short size bond
paper inside a short plastic envelop with your names and submit them in designated
pick-up areas.
I hope that you will find this course interesting and fun. I hope to know more of your
experiences, insights, challenges and difficulties in learning as we go along this course.
I am very positive that we will successfully meet the objectives of the course.
May you continue to find inspiration to become a great professional. Keep safe and God
bless!
Course
EDMM113
Number
Course Title Calculus 3
This course aims to provide students with an understanding of the applications of
Course differentiation and integration in sequences, infinite series, power series, as well as multiple
Description integration for functions in several variables. Moreover, students will be able apply these
concepts to problem solving.
No. of Units 3 units
Pre-requisites None
1. Exhibit in-depth understanding of the theories in differentiation and integration of
sequences, infinite series, power series, as well as multiple integration for functions
Course containing two or more variables.
2. Apply the theory in solving problems related in sequences, infinite series, power series, as
Intended
well as multiple integration for functions containing two or more variables.
Learning 3. Demonstrate mathematical skills in solving problems involving differentiation and
Outcomes integration of sequences, infinite series, power series, and multiple integration for functions
on several variables.
4. Develop a work attitude of doing things correctly, honestly and independently.
MODULE 1
Lesson 1: Sequence
Lesson 2: Limit of Sequence
Lesson 3: Types of Sequence
MODULE 2
Content
Coverage Lesson 4: Types of ConvergenceTest
MODULE 3
Lesson 5: Types of Infinite Series
Lesson 6: Application of multiple integration
References REFERENCES:
TEXTBOOK:
T1: Anton, Howard (1988). Calculus with Analytic Geometry, Third Edition by Anton
Textbooks, Inc.
ONLINE REFERENCES:
OR1. exponential function https://www.youtube.com/watch?v=oo1ZZlvT2LQ
OR2. inverse function: https://www.youtube.com/watch?v=I_ril7ToAi4
or any part of this module is not authorized without the permission from
MODULE
2
LESSO
N
In this module, we will discuss about the different types of convergence tests in an infinite series. There
are basic concepts about infinite series that we need to understand while looking forward to apply these
various tests of convergence.
- define (a) infinite series, (b) sum of infinite series, (c) partial sum of infinite series
In this section, we will study about the different types of convergence tests. The following are the types
of convergence tests and their respective usage.
We decided to present the details of each convergence test in a table containing with the comments or
usage of each test.
Integral Test Let ∑ u k be a series with Use this test when f(x) is easy to
positive terms and let f(x) be the integrate.
function that results when k is This test only applies to series
replaced by in the formula for uk. that have positive terms.
if f is decreasing and continuous
∑ u k and ∫ f ( x ) dx
k =1 1
Comparison Test Let ∑ ak and ∑ bk be series Use this test as a last resort.
with positive terms such that Other tests re often easier to
apply.
a1 ≤ b1, a2≤ b2, . . . , ak ≤ bk, . . .
This test only applies to series
If ∑ bk converges, then ∑ ak with positive terms.
converges, and if ∑ ak diverges,
then ∑ bk diverges.
Ratio Test Let ∑ u k be a series with Try this test when uk involves kth
positive terms and suppose factorial or powers.
u k+1
lim =ρ
k →+∞ uk
(a) Series convergences if
ρ<1.
(b) Series diverges if ρ>1 or
ρ = +∞ .
(c) No conclusion if ρ=1.
Limit Comparison Test Let ∑ ak and ∑ bk be series This is easier to apply than the
with positive terms such that comparison test, but still
requires some skill in choosing
ak the series ∑ bk for comparison.
ρ = lim
k→∞ bk
If 0 < ρ < +∞ , then both series
converge and both diverge.
converge if
(a) a1 ≥ a2 ≥ a3 ≥ . . .
(b)
Ratio Test for Absolute Let ∑ u k be series with nonzero The series need not have
Convergence terms such that positive terms and need not be
alternating to use this test.
|u k+1|
ρ=lim
k →∞ |uk|
Root Test Let ∑ u k be a series with Try this test when uk involves kth
positive terms and suppose power
ρ = lim √ uk = klim
k 1 /k
(u k )
k→∞ →∞
Before we will discuss the different types of convergence tests, let us familiarize first these important
concepts which are the foundation of the study in this entire subject.
∑ uk
k =1
+∞
The number sn is called the nth partial sum of the series and the sequence { sn }n=1 is called the sequence
of partial sums.
3
S1 =
10
3 3 33
S2 = + 2=
10 10 100
3 3 3 333
S3= + 2+ 3=
10 10 10 1000
3 3 3 3 3333
S4 = + 2+ 3+ 4=
10 10 10 10 10000
.
.
.
As n increases, the partial sum sn = u1 + u2 + u3 + . . . + un includes more and more terms of the series.
Thus, if sn tends toward a limit as n→+∞ , it is reasonable to view this limit of is the sum of all the terms
in the series.
n
Definition 1.1.2: Let { sn } be the sequence of the partial sums of the series ∑ u k . If the sequence { sn }
k =1
converges to a limit S, then the series is said to converge and S is called the sum of the series. We
denote this by writing
If the sequence of the partial sum diverges, then the series is said to diverge. A divergent series has no
sum.
1 3 3 3 3
= + 2 + 3 +...+ k +...
3 10 10 10 10
Let us verify that this is indeed the case. The nth partial sum is
3 3 3
Sn = + 2 +...+ n (1)
10 10 10
1 3 3 3 3
Sn = 2 + 3 + . . . + n + n+1 (2)
10 10 10 10 10
and then subtracting the new expression (2) from the previous (1) to obtain:
1 3 3
Sn - sn= - n+1
10 10 10
9
10
sn=
3
10
1
1− n
10 ( )
sn=
1
10 (
1
1− n
10 ) (3)
1 1
Since n → 0 as n →+∞ , it follows from (3) that the S =
lim s n= .
10 n →+∞ 3
Thus,
1 3 3 3 3
= + 2 + 3 +...+ n +...
3 10 10 10 10
1.2 Convergence
Our first theorem states that the terms of an infinite series must tend toward zero if the series is to
converge.
uk = sk – sk – 1
where sk is the sum of the first k terms and sk – 1 is the sum of the first k-1 terms. If S denotes the sum of
the series, then klim s k−1=S, and since (k-1) →+∞ as k →+∞ , we also have lim s k−1=S. Thus,
→+∞ k →+∞
∞
k 1 2 3 k
∑ k +1 = + + +...+
2 3 4 k +1
+...
k =1
diverges since
1
k lim =¿ ¿
lim =¿ ¿ k →+∞ 1 1 ≠ 0.
k →+∞ k+1 1+
k
Warning: The converse of theorem 4.2.1 is false. To prove that a series converges it does no suffice to
show that klim uk =0 , since this property holds for divergent as well as convergent series. For example,
→+∞
1 1 1
the kth term of the divergent harmonic series 1 ++ + . . . + + . . . tends to zero as k →+∞ , and
2 3 k
1 1 1
the kth term of the convergent geometric series + 2 + . . . + k + . . . tends to zero as k →+∞ .
2 2 2
∑ ( uk +v k ) = ∑ u k + ∑ v k
k =1 k =1 k =1
∞ ∞ ∞
∑ ( uk −v k ) = ∑ u k - ∑ v k
k =1 k =1 k =1
b. If c is a nonzero constant, then the series ∑ u k and ∑ cuk both converge or both diverge. In
the case of convergence, the sums are related by
∞ ∞
∑ cuk = c ∑ u k
k =1 k =1
∑ uk = u 1 + u2 + u3 + …
k =1
and
∑ uk = u k + uk+1 + uk+2 + …
k= K
( 43 − 52 )
∞
∑ k k−1
k =1
∞
3 3 3
∑ 43k = 4 + 4 2 + 4 3 . . . + . . .
k =1
3 1
is a convergent geometric series (a = , r = ), and the series
4 4
∞
2 2 2 2
∑ 5k−1 =2 + + +
5 52 53
+...
k =1
3
2
( )
∞ ∞ ∞
3 2 3 2 4 5 3
∑ k
4 5
− k−1 = ∑ ( k¿
)− ∑ ( k−1 ¿
) ¿ ¿ = -
1 1−
1 =1- =-
2 2
k =1 k =1 4 k=1 5
1− 5
4
∞
5 5 5
∑ 5k = 5 + 2 + 3 + … + k + …
k =1
()
∞ ∞
∑ 5k = ∑ 5 1k
k =1 k =1
so each term is constant times the corresponding term of the divergent harmonic series.
∞
1 1 1 1
∑ k
=
10
+
11
+
12
+…
k=10
diverges by part (c) of Theorem 4.2.3, since this series results by deleting the first nine terms from the
divergent harmonic series.
Convergence Tests
If an infinite series u1 + u2 + u3 + … + has positive terms, then the partial sums s 1 = u1, s2 = u2, s3 = u3, …
form of an increasing sequences, that is
s1 < s2 < s3 < … < sn < …
If there is a finite constant M such that s n ≤ M for all n, then the sequence of partial sums will converge
to a limit S satisfying S ≤ M. If no such constant exists, then nlim
→∞
s n= +∞ . This yields the following
theorem.
Theorem 1.3
If ∑ u k is a series with positive terms, and if there is a constant M such that
for every n, then the series converges and the sum S satisfies ≤ M. if no such M exists then the series
diverges.
∑ k12
k =1
+∞
∫ x12 dx
1
whose integrand is obtained by replacing the summative index k by x, then there is a relationship
between convergence of the series and convergence of the improper integral.
Integral Test
Theorem 1.4
Let ∑ u k be a series with positive terms and let f(x) be the function that results when k is replaced by x in
the formula for uk. If f is decreasing and continuous for x≥ 1, then
∞ +∞
∑ u k and ∫ f ( x ) dx
k =1 1
∑ k12
k =1
converges or diverges.
1
f(x) = 2
x
which satisfies the hypotheses of the integral test. (Verify) Since
[ ] [ ]
+∞ l l
1 dx −1 1
∫ x
2
dx = lim ∫ 2 = lim
l→∞ 1 x l→∞ x 1
= lim ¿l → ∞ 1− ¿ = 1
l
1
∫ x12 dx=1. (If you write out the terms of the series, it will be evident that the sum exceeds 1.)
1
Example 6. The integral test provides another way to demonstrate divergence of the harmonic series
∞
1
∑ 1k . If we replace k by x in the formula u k we obtain the function f(x) =
k
, which satisfies the
k =1
hypotheses of the integral test. (Verify.) Since
+∞ l
1 1 [ lnl−ln1 ] = + ∞
∫ x
dx = lim ∫ dx = lim
l→∞ 1 x l→∞
1
1 2 3 k
+ 4 + 9 +…+ k +… 2
e e e e
converges or diverges
Solution:
x
f ( x )= x
2 = xe−x
2
e
For x ≥ 1, this function has positive values and is continuous. Moreover, for x ≥ 1 the derivative
is negative, so that f is decreasing for x ≥ 1. Thus, the hypotheses of the integral test are met. But
[ ]
+∞ l l
−1 −x
∫ xe dx=lim ∫ xe−x dx = lim
2 2
−x 2
e
1 l →∞ 1 l→∞ 2 1
( ) −1
lim [ e−l −e−1 ] =
1
2
=
2 l→∞ 2e
u k+1
lim =ρ
k → ∞ uk
∑ k1!
k =1
u k+1 1 / ( k +1 ) ! k! 1
ρ = lim = lim = lim = lim =0
k → ∞ uk k→∞ 1/k ! k → ∞ ( k +1 ) ! k → ∞ +1
k
so that ρ < 1.
∑ 2kk
k =1
u k+1 k
k +1 2 1 k +1 1
ρ = lim = lim k +1 . = lim =
k → ∞ uk k→∞ 2 k 2 k→∞ k 2
so that ρ < 1.
u k+1 ( k +1 )k +1 k !
ρ = lim = lim . k
k → ∞ uk k → ∞ ( k+ 1 ) ! k
( k +1 )k
= lim
k→∞ kk
( )
k
1
= lim 1+
k→∞ k
=e
1 1 1 1
1+ + + +...+ +...
3 5 7 2 k−1
converges or diverges.
However, the integral test proves that the series diverges since
+∞ l
dx dx ¿ln ( 2 x −1 ) ] l = +∞
∫ 2 x−1 = lim ∫ = llim 1
l →+∞ 1 2 x−1
→+∞
1
2! 4 ! 6 ! (2 k )!
+ 2 + 3 +...+ +...
4 4 4 4k
diverges since
u k+1 [2 ( k +1 ) ! ] 4k (2 k +2)! 1
ρ = lim = lim . = lim .
k → ∞ uk k→∞ 4
k +1
(2 k )! k →∞ (2 k )! 4
1 lim ¿
= (2k + 2)(2k + 1) = +∞
4 k→∞
ρ = lim √ uk = klim
k 1 /k
(u k )
k→∞ →∞
∑ ( 42 k−5
k+1 )
∞ k
k =1
(u k )
1 /k 4 k −5
ρ = klim = lim =2>1
→∞ k →+∞ 2 k +1
lim (u k )
1 /k 1
= lim =0<1
k→∞ k →+∞ 2 k +1
Comments on Notation
We conclude this section with a remark about notation. Until now we have written most of our infinite
series in the form
∞
∑ uk (4)
k =1
with the summation index beginning at 1. If the summation index begins at some other integer, it is
always possible to rewrite the series in the form (4). Thus, for example, the series
Can be written as
∞
2k−1 2
2
2
3
∑ ( k −1 ) ! = 1 + 2 + 2! + 3! + . . . (6)
k =1
However, for purposes of applying convergence tests, it is not necessary that the series have form (4).
For example, we can apply the ratio test to (5) without converting to the more complicated form (6).
Doing so yields
u k+1 2
k+1
k! 2
ρ = lim = lim . k = lim =0
k → ∞ uk k → ∞ ( k +1 ) ! 2 k → ∞ k +1
which shows that the series converges since ρ < 1.
Theorem 4.6.1 Let ∑ ak and ∑ bk be the series with positive terms and suppose
a 1 ≤ b1 , a 2 ≤ b2 , a 3 ≤ b3 , . . . , a k ≤ b k , . . .
(a) If the “bigger series” ∑ bk converges, then the “smaller series” ∑ ak also converges.
(b) On the other hand, if the “smaller series” ∑ ak diverges, then the “bigger series” ∑ bk also
diverges.
There are two basic steps required to apply the comparison test to a series ∑ u k of positive terms:
2. Find a series that proves a guess to be correct. Thus if the guess is convergence we must find a
divergent series whose terms are “smaller” than the corresponding terms of ∑ u k , and if the guess is
convergence we must find a convergent series whose terms are “bigger” than the corresponding terms
of ∑ u k .
Below is the principle used in making a guess in step 1. There is no guarantee that this principle works
always but it works often enough to be useful as a starting point for the comparison test.
Constant terms in the denominator of uk can usually be deleted without affecting the convergence or
divergence of the series.
Example 1. Use the above principle to guess whether the following series converge or diverge.
The modified series is a convergent geometric series so the given series is likely to converge.
1
The modified series is a portion of a divergent p-series (p = ), so the given series is likely to diverge.
2
1
c. Deleting the suggests that
2
∞
1
∑ behaves like ∑
∞
1
( )
3
k =1 1
k+ k =1 k3
2
The modified series is a convergent p-series (p=3) so the given series is likely to converge.
If a polynomial in k appears as a factor in the numerator or denominator of uk, all but the highest power
of k in the polynomial may usually be deleted without affecting the convergence or divergence of the
series.
Example 2. Use the above principle to help guess whether the following series converge or diverge.
∞ ∞
1 6 k 4−2 k 3 +1
(a) ∑ (b) ∑
k =1 √ k 3 +2 k k =1 k 5+ k 2−2 k
Solution:
Since the modified series is a constant times the divergent harmonic series, the given series is
likely to diverge.
Once it is decided whether a series is likely to converge or diverge, the second step in applying the
comparison test is to produce a series with which the given series can be compared to substantiate the
guess. Let us consider the case of convergence first. To prove ∑ ak converges by the comparison test
we must find a convergent series ∑ bk such that
ak ≤ bk
for all k. Frequently, b k is derived from the formula for a k by either increasing numerator of a k , or
decreasing the denominator of a k or both.
∑ 2 k12 +k
k =1
converges or diverges.
Solution. Using the principle 4.8.2, the given series behaves like the series
∞ ∞
∑ 21k 2 = 12 ∑ k12
k =1 k=1
Which is a constant times a convergent p-series. Thus, the given series is likely to converge. To prove the
1
convergence, observe that when we discard the k from the denominator of 2 , the denominator
2k +k
decreases and the ratio increases, so that
1 1
2 < 2
2k +k k
For k = 1, 2, 3, . . . .
Since
∑ 21k 2 = 12 ∑ k12
k =1 k=1
converges, so does
∞
∑ 2 k12 +k
k =1
by comparison test.
∑ 2 k 12−k
k =1
converges or diverges.
Solution. Use principle 4.8.2, the series behaves like the convergent series
∞ ∞
∑ 21k 2 = 12 ∑ k12
k =1 k=1
1
Thus, the given series is likely to converge. However, if we discard k from the denominator of 2 ,
2 k −k
the denominator increases and the ratio decreases, so that
1 1
2 > 2
2 k −k 2 k
Unfortunately, this inequality is in the wrong direction to prove convergence of the given series. A
different approach is needed; we must do something to decrease the denominator, not increase it. We
accomplish this by replacing k by k2 to obtain
1 1 1
2
≤ 2 2= 2
2 k −k 2 k −k k
1
= 2
k
∞
1
Since ∑ is a convergent p-series, the given series converges by the comparison test.
k =1 k2
To prove that a series ∑ ak diverges by the comparison test, we must produce a divergent series ∑ bk
of positive terms such that a k ≥ b k for all k.
Solution. Use the principle 4.8.1, the series behaves like the divergent harmonic series
∞
∑ 1k
k =1
1
1
1 > for k = 1, 2, . . .
k− k
4
∞
1
and since ∑ diverges, the given series diverges by comparison test.
k =1 k
∑ √ k1+5
k =1
converges or diverges.
Solution. Using principle 4.8.1 the series behaves like the divergent p-series
∞
∑ √1k
k =1
1 1 1
≥ =
√ k +5 √ k + √ k 2 √k
and since
∞
∑ 2 √1 k
k=25
∑ √ k1+5 diverges by the comparison test; consequently, the given series diverges by theorem 4.2.3
k=25
(c) Algebraic Properties of Infinite Series.
Theorem 4.9.1. Let ∑ ak and ∑ bk be the series with positive terms and suppose
ak
ρ=lim
k →∞ bk
If ρ is finite and ρ ≠ 0, then the series both converges or both diverges.
To illustrate that this test works we shall use the same expressions given in examples 4 and 5.
Example 7. Use the limit comparison test to determine whether the following series converge or
diverge.
∞
1
(b) ∑
∞
1
(a) ∑ 2 k =1 1
k=1 2 k −k k−
4
Solution:
(a) In example 4, we guess that the given series behaves like the convergent series
∞
∑ 21k 2
k =1
1 1
ak = 2 and b k = 2
2 k −k 2k
we obtain
1
ak 2 2 2
2 k −k 2k lim
ρ=lim = lim = lim = k→∞ 2 1 =1
k →∞ b k k→∞ 1 2
k → ∞ 2 k −k 2−
k
2 k2
∞
1
Since∑ converges, so does the given series since ρ is finite and positive.
k =1 2 k2
(b) In example 5, we guess that the given series behaves like the divergent series
∞
∑ 1k
k =1
1
1
ak = 1 and b k =
k− k
4
a k lim k lim
1
ρ=lim = k→∞ 1 = k→∞ 1 =1
k →∞ bk k− 1−
4 4k
∞
1
Since ∑ diverges, so does the given series since ρ is finite and non-zero.
k =1 k
Example 8. Use the limit comparison test to determine whether
∞ 3 2
∑ 3 kk 5−23
k +4
−k +2
k =1
converges or diverges.
∑ 3kk5 = ∑ k32
k =1 k =1
which converges since it is a constant times a convergent p-series. Thus the given series is likely to
∞ 3 ∞
3k 3
converge. To substantiate this, we apply the limit comparison test to series ∑ 5
= ∑ 2 and the
k =1 k k =1 k
given series. We obtain
3 2
3 k −2 k + 4
ak 5 3
k −k + 2
5 4
3 k −2 k + 4 k
2
ρ=lim = lim = lim 5 3 =1
k →∞ b k k→∞ 3 k → ∞ 3 k −3 k +6
2
k
∞ 3 ∞
3k 3
Since ρ ≠ 0, the given series converges because the series ∑ 5
= ∑ 2 converges.
k =1 k k =1 k
The previous series contains only positive terms. In this section we will discuss series containing negative
terms.
Alternating Series. It is a series whose terms are alternatively positive and negative. These are called
alternating series. Such series have one of two possible forms:
a 1 - a 2 + a 3 - a 4 + . . . + (−1 )k +1 a k + . . .
Or
−a 1 + a 2 - a 3 + a 4 - . . . + (−1 )k +1 a k + . . .
∑ (−1 ) k+1
ak or ∑ (−1 ) ak
k
k =1 k =1
(a) a 1 ≥ a 2 ≥ a3 ≥ . . . ≥ a k ≥ . . .
(b) klim a k =0
→∞
1 1 1
1− + - + . . . + (−1 )k +1 a k + . . .
2 3 4
is called alternating harmonic series. Since
1 1
ak = > = a k+1
k k +1
and
lim a k =¿ ¿ lim 1 =¿ ¿ 0
k→∞ k→∞ k
converges or diverges.
1 3
+
lim a k =¿ ¿ lim k +3 k k2
=¿ ¿ lim =¿ ¿ 0
k→∞ k → ∞ k (k + 1) k→∞ 1
1+
k
to see if requirement (a) is met, we must determine whether the sequence
is non-increasing. Since
ak +1 k+4 k (k +1) 2
k +4k 2
k +4 k
= . = 2 = <1
ak (k + 2)(k +1) k + 3 k +5 k +6 (k ¿¿ 2+ 4 k)+(k +6)¿
we have a k > a k+1, so the series converges by alternating series test.
Remark. If an alternating series violates condition (b) of the alternating series test, then the series must
diverge by the divergence test. However, if condition (b) is satisfied, but (a) is not, the series may either
converge or diverge.
1.10.2 Theorem. If an alternating series satisfies the hypotheses of the alternating series test, and if
the sum S of the series is approximated by the nth partial sum sn , then the absolute value of the error
is less than or equal to a n+1.
To give us the idea about this theorem, we presented the figure below. Figure 4.10.1 provides some
insight to the way in which an alternating series
a 1 - a 2 + a 3 - a 4 + . . . + (−1 )k +1 a k + . . .
converges to its sum S when the hypotheses of the alternating test are satisfied.
a2
a3
a4
a5
_.___.____.___.____._
s2 s4 S s5 s3 s1
Figure 4.10.2
In the figure we have plotted the successive partial sums on the x-axis. Because
a1 ≥ a2 ≥ a3 ≥ a4 ≥ . . .
and klim a k =¿ ¿ 0
→∞
the successive partial sums oscillate in smaller and smaller steps, closing in on the sum S. It is of interest
to note that the even-numbered partial sums are less than or equal to S and the odd-numbered sums
are greater than or equal to S. thus the sum S falls between any two successive partial sums; that is, for
any positive integer n
|S−s n| ≤ |S n+1−sn|
But Sn +1−s n = ± an +1 (the sign depending on whether n is even or odd) so it yields
|S−s n| ≤ ± an +1
Since |S−s n| represents the magnitude of error that results when we approximate the sum of the
entire series by the sum of the first n terms,
|S−s n| ≤ ± an +1
tells u that this error is less than or equal to the magnitude of the (n+1)-st term in the series. The reader
can check that
|S−s n| ≤ ± an +1
also holds for alternating series of the form
−a 1 + a 2 - a 3 + a 4 - . . . + (−1 )k +1 a k + . . .
1 1 1
1− + - + . . . + (−1 )k +1 a k + . . .
2 3 4
satisfies the hypotheses of the alternating series test. If we appropriate the um of the series by
1 1 1 7
1− + - =
2 3 4 12
1
then the absolute value of the error is at most = 0.2, and if we approximate the sum of the series by
5
1 1 1 1 1 1 319
1− + - + - + =
2 3 4 5 6 7 420
1
Then the absolute value of the error is a t most = 0.125.
8
The series
1 1 1 1 1 1
1- - + + - - +...
2 22 23 24 25 26
does not fit in any of the categories studied so far – it has mixed signs but is not alternating. We shall
now develop some convergence tests that can be applied to such series.
converges.
1 1 1 1 1 1
1- - + + - - +...
2 22 23 24 25 26
is a convergent series . on the hand, the alternating harmonic series
1 1 1 1
1− + - + - . . .
2 3 4 5
does not converge absolutely since the series of absolute values
1 1 1 1
1+ + + + + . . .
2 3 4 5
diverges.
1 1 1 1 1 1
1- - + + - - +...
2 22 23 24 25 26
converges absolutely. It follows from Theorem 1.11.2 that the series converges.
∑ cosk 2 k
k =1
converges.
| |
cos k
k 2
1
≤ 2
k
Thus,
∞
∑ cosk 2 k
k =1
converges.
1 1 1 k +1 1
1− + - + . . . + (−1 ) +... (1)
2 3 4 k
1 1 1 1
-1− - - - ...- -... (2)
2 3 4 k
Series (1), the alternating harmonic series, converges; while series (2), being a constant times the
harmonic series, diverges. Yet in each case the series of absolute values is
1 1 1
1+ + + ...+ +...
2 3 k
which diverges. A series such as (1), which is convergent, but not absolute convergent, is called
conditionally convergent.
|u k+1| 2
k+1
k! 2
ρ=lim = lim . k = lim =0<1
k →∞ |uk| k → ∞ ( k +1 ) ! 2 k → ∞ k +1