Unit 9 Alternating Series

UNIT 9

ALTERNATING SERIES

Structure Page Nos.

Nos.
9.1 Introduction 263
Objectives
9.2 Alternating Series 264
Leibniz’s Test
9.3 Absolute and Conditional Convergence 268
9.4 Rearrangement of Series 270
9.5 Summary 274
9.6 Solutions/Answers 275

9.1 INTRODUCTION
In the previous unit we discussed some special tests for checking the
convergence or divergence of positive term series. You know that an infinite
series need not always be a positive term series. In fact, infinite series, in
general, can have both an infinite number of negative terms as well as an
infinite number of positiver terms. The series which have both negative and
positive terms may be classified into two major categories. The first category
consists of those infinite series whose terms are alternately positive and
negative. Such series are called Alternating Series. The other category is
one in which terms need not necessarily be alternatively positive and negative
that is to say, the infinite series whose terms are mixed and do not follow any
specific pattern.

The question, now arises: How to test the convergence of an infinite series
which has both positive and negative terms? The convergence tests
discussed in Unit 8 are not suitable for this purpose because these tests in
their present form cannot be applied to these series. Accordingly, the whole
content of the unit is divided into three sections.

In Sec. 9.2, we discuss alternating series and its convergence. We shall

discuss an important test known as Leibniz Test which gives sufficient
conditions for convergence of an alternating series.

In Sec. 9.3, we introduce two types of convergence for an alternating series;

namely absolute and conditional. You will see that all the tests for checking
the convergence of positive term series are useful for checking the absolute
convergence and the conditional convergence may hold when the absolute 263
Block 3 Infinite Series
convergence fails.

Finally, in Sec. 9.4 we briefly discuss the method of rearrangement of series.

Objectives
Therefore, after studying this unit, you should be able to

• recognize an Alternating Series;

• apply the Leibniz Test to check the convergence of an Alternating series;
• identify an absolutely convergent series and a conditionally convergent
series;
• generate new series by rearranging terms of an infinite series and check
the nature of the convergence or divergence of the new series.

9.2 ALTERNATING SERIES

In this section we turn our attention to series that have positive as well as
negative terms. Of these, the simplest are those whose terms are alternatively
positive and negative.

Some of the examples of an alternating series are

1 1 1 1
− + − + ...,
3 5 7 9
1⋅ 2 1⋅ 2 ⋅ 3 1⋅ 2 ⋅ 3 ⋅ 4
1− + − + ...,
1⋅ 3 1⋅ 3 ⋅ 5 1⋅ 3 ⋅ 5 ⋅ 7
and 1 − 1 + 1 − 1 + ...

Formally, we define an alternating series in the following way:

∞
Defintion 1: An infinite series ∑a
n =1
n is called an alternating series if any two

consecutive terms of the series are of opposite sign.

An alternating series can be written as

∑ (−1)
n =1
n +1
an = a1 − a2 + a3 − a4 + ...., where each an > 0 . ... (1)

If the first term is negative, it can be written as

∑ (−1) n
an = − a1 + a2 − a3 + a4 + .... ... (2)

The second series can be obtained from the first if you multiply each term of
the first series by (−1) . Therefore, it is enough to discuss the convergence of
the first series.

Note that the terms of an alternating series are alternatively positive and
negative. The terms with positive signs are called positive terms and the terms
with negative sign are called negative terms.

264
Unit 9 Alternating Series
There is a very simple test for the convergence of an Alternating series. This
test is known as Leibniz Test after the name of Leibniz, the eminent German
mathematician.
∞
Theorem 1 (Leibniz Test): Let ∑ (−1)
n =1
n+1
an be an alternating series such that

i) an > 0 ∀ n = 1, 2, 3, K

ii) a1 ≥ a2 ≥ a3 ≥ K , i.e., (an ) n is a monotonically decreasing sequence.

iii) lim an = 0
n →∞

∞ ∞
Then the series ∑ (−1) n+1 an is convergent. Moreover, if s = ∑ (−1) n−1 an and
n =1 n =1

sn is the nth partial sum, then s − s n < a n+1 .

Gottfried Leibniz
(1646-1716)
Proof: Let us first look at the even partial sums s 2 , s 4 , s6 ,... We have
Leibniz Test is also
s2 = a1 − a2 ≥ 0, Known as Alternating
Series Test.
s4 = s2 + (a3 − a4 ) ≥ s2 because a3 ≥ a4 ,
s6 = s4 + (a5 − a6 ) ≥ s4 because a5 ≥ a6 ,

and, in general,

s2 n+ 2 = s2 n + (a2 n+1 − a2n+ 2 ) ≥ s2 n because a2 n+1 ≥ a2 n+ 2

Then s 2 ≤ s4 ≤ s6 ≤ ...... … (3)

Also s 2 n = a1 − (a2 − a3 )...(a2 n−2 − a2 n−1 ) − a2 n .

It follows that s 2 n ≤ a1 ∀ n ∈ N. Monotonic convergence then implies that the

subsequence (s 2 n ) converges to some s. We next show that ( sn ) converges to
1 1
s. Indeed, if ε > 0 ∃ K such that n ≥ K ⇒ s2 n − s ≤ ε, and a2n+1 ≤ ε .
2 2

Therefore s2 n +1 − s ≤ s2 n − s + a2 n +1 ≤ ε.

Thus every partial sum of odd terms is within ε of s. Since ε > 0 is arbitrary
( sn ) converges to s and hence ∑ (−1) n +1
⋅ an is convergent and

∑ (−1)
n+1
an = s.

Now we check the inequality s − s n < a n +1 .

we note that ( −1) n + 2 an +1 + ... + ( −1) n + p +1 an + p = an +1 − an + 2 + ... + a p −1 an +9

The sum between absolute value signs can be expressed in the form

(an+ p−1 − an+ p ) of p is even

(an+1 − a n+ 2 ) + (an+3 − an+ 4 ) + ... + 
an+ p if p is odd 265
Block 3 Infinite Series
Since ( an ) is decreasing this shows that the sum is ≥ 0 and therefore the
absolute value sign on the right above can be removed. The sum is can be
written as

an+ p if p is even
an+1 − ( a n+ 2 − a n−3 ) + ...
( an+ p−1 − a n+ p ) if p is odd

This shows that the sum is ≤ a n +1 .

Hence the inequality s − s n < a n +1 holds.

Hence the theorem.

From Unit 7, you know that the condition lim an = 0 is necessary for the
n →∞
∞
convergence of the series ∑a
n =1
n . But according to Leibniz Test, if the given

series is an alternating series where the absolute value of the terms form a
monotonically decreasing sequence, then the condition lim an = 0 is also
n →∞
sufficient for the convergence.

Let us do some examples and exercises:

Example 1: Using the Leibniz Test show that the following series converges.

∞
(−1) n+1 1 1 1 1
∑
n =1 n
= 1 − + − + ......
2 3 4 5

Solution: This series is known as the Alternating Harmonic Series.

Let us check whether the conditions of the Leibniz Test are satisfied.

∞
1
We can write the series as ∑ (−1)
n =1
n +1
an , where an =
n
.

Here

i) an > 0 for all n.

ii) a1 > a2 > a3 > a4 > .......
1
iii) lim an = lim = 0 .
n →∞ n →∞ n

Thus, the series satisfies all the conditions of the Leibniz Test.

Hence, the series is convergent.

Can you guess what the sum of this series is?

It was proved as a well known result that the limit or sum of this series is
equal to ln 2 .
***
Example 2: Check whether the following series satisfies all the conditions of
the Leibniz Test.
266
Unit 9 Alternating Series
1 1 1 1
2 3 4 5
3−3 +3 −3 +3 L
1 ∞
Solution: Let an = 3 n . Then the series can be written as ∑ (−1)
n =1
n +1
an . Since
1 1
3 > 3 2 > 3 3 > K , therefore the first and second conditons of the Leibniz test
are satisfied.
1
n
However, lim an = lim 3 = 1 ≠ 0 . (Recall from Unit 6)
n →∞ n →∞

That means, the third condition of the Leibniz test is not satisfied.
***
Example 3: Show that the series

∞
2n + 1 3 5 7 9
∑ (−1)
n =1
n −1
= − + −
n(n + 1) 2 2.3 3.4 4.5
+ ...
is convergent.
2n + 1
Solution: For the given series we have, an = . Clearly, an > 0 and
n(n + 1)
lim an = 0. Now
n →∞

an+1 2n + 3 n(n + 1) 2n 2 + 3n
= ⋅ = 2 <1
an (n + 1) (n + 2) 2n + 1 2n + 5n + 2

Hence an > an+1 , and therefore the series converges.

***
Now see if you can solve these exercises.

1 1 1
E1) Show that the series 1 − + − + ... is convergent.
2! 3! 4!

1 1 1 1
E2) Show that the series p
− p + p − p + L converges for p > 0.
1 2 3 4

Here we make an important point. The first step in applying the Alternating
Series Test is to check that the terms actually alternate in sign. Clearly, the
series

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1− + + + + + + + + + + + + + + + − ...
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
has positive and negative terms; yet, it is not an alternating series and
therefore the Alternating Series Test cannot be applied. The series diverges.
Nor does the alternating series

e − e1/ 2 + e1/ 3 − e1/ 4 + ...

meet the requirements of the Alternating Series Test, even though an = e1/ n is 267
Block 3 Infinite Series
positive and decrease with n. The difficulty is that lim an ≠ 0 (Since as
n→∞
1
n
n→∞ e = 1. ) Thus the series diverges [Refer Theorem 1 in Unit 7].

Let us now consider once again the Harmonic Series

∞
1 1 1
∑ n = 1+ 2 + 3 +L
n =1

You know that this series diverges. But the Alternating Harmonic Series (as
discussed in Example 1) namely

∞
1 1 1 1
∑ (−1)
n =1
n+1

n
=1− + − +L
2 3 4

converges. Thus, we have a series that converges because it is an alternating

series. If all the negative terms are replaced by the corresponding positive
terms, then the convergence is lost. To study this phenomenon in a more
general way, we introduce the notions of absolute convergence and
conditional convergence in the next section.

9.3 ABSOLUTE AND CONDITIONAL

CONVERGENCE
In this section we shall discuss two types of convergence of an altenating
series namely absolute convergence and conditional convergence.

Let us consider the following two series:

1 1 1 1
i) 1− + − + L
2 4 8 16
and

1 1 1 1
ii) 1− + − + L
2 3 4 5

By the Leibniz Test, both the series converge.

Again consider the two series

1 1 1 1
iii) 1+ + + + L and
2 4 8 16
1 1 1 1
iv) 1+ + + + L
2 3 4 5

obtained from (i) and (ii) by replacing each negative term by its absolute value
(i.e − k to + k ) . Then you know from Unit 7 that the series (iii) converges,
while the series (iv) diverges. This shows that if you replace an alternating
series with a series with all terms having positive signs, then in some cases
we get convergence whereas in other cases divergence. This leads us to
divide the convergent series into two classes, namely, the ‘absolutely
convergent series’ and the ‘conditionally convergent series’, which we define
as follows:
268
 Unit 9 Alternating Series
∞
Definition 2 (Absolute convergence): A series ∑a n =1
n of real numbers is said
∞
to be absolutely convergent if ∑a
n =1
n is convergent.

. ∞
Definition 3 (Conditional Convergence): A series ∑a
n =1
n of real
∞
numbers is said to be conditionally convergent if ∑a
n =1
n converges and
∞

∑a
n =1
n does not converge i.e. the given series is convergent but not absolutely

convergent.

For example the series in (i) in the paragraph at the beginning converges
absolutely, while the series (ii) there converges conditonally.

∞ ∞
Note: We have defined ∑ an to be absolutely convergent if
n =1
∑| a
n =1
n | is
∞
convergent, but we have not said anything about the behaviour of ∑a
n =1
n itself.
∞ ∞
Whereas we have defined ∑a
n =1
n to be conditonally convergent if ∑a
n =1
n is
∞
convergent but ∑| a
n =1
n | is divergent.

Next we shall prove another theorem.

Theorem 2: If an infinite series is absolutely convergent, then it is convergent.

Proof: Let ∑a n be an absolutely convergent series i.e., ∑| a n | is

convergent. We have to prove that ∑a n is convergent.
Let ( sn ) be the sequence of partial sums of ∑a n . Then

sn = a1 + a2 + L + an .

It is enough to show that (sn ) n is a Cauchy sequence.

Let (t n ) n be the sequence of partial sums associated with the series ∑| a n |.

Since ∑| a n | is convergent, therefore (t n ) n is convergent. Thus (t n ) n is a
Cauchy sequence. In other words, for an ε > 0 , there exists a positive integer
m such that

| t n − t k | < ε for n > m, k > m .

Suppose n > k . Then

| sn − sk | = | ak +1 + ak +2 + L + an |
≤ | ak +1 | + | ak + 2 | + L + | an | (Recall from Unit 3).
= | tn − tk | < ε . 269
Block 3 Infinite Series
Which shows that (sn ) n is a Cauchy sequence. This completes the proof.

Thus every absolutely convergent series is convergent. The converse,

however, is not true. That is to say if a series is convergent, then it may not be
absolutely convergent. The following example shows this.

Example 4: Test the absolute and conditional convergence of the series

∞
(−1) n+1
∑
n =1 2n + 1
.

∞ ∞
1
Solution: Here ∑ | an | = ∑
n =1 n =1 2n + 1
∞ ∞
1 1
Let bn =
n
for n = 1, 2, 3, K Then the series ∑ bn = ∑
n =1 n =1 n
is divergent.

Also
| an | n 1
lim = lim = .
n→∞ b n→∞ 2n + 1 2
n

∞
Hence by the Comparison Test, it follows that ∑| a
n =1
n | is divergent. Thus
∞
(−1) n+1
∑
n =1 2
n +1
is not absolutely convergent.

1 1 1
However, since < ∀ n , and lim = 0 , therefore by the
2n + 1 2n − 1 n →∞ 2n + 1
∞
(−1) n +1
Leibniz Test, ∑ is convergent. Hence the series is conditionally
n =1 2n + 1
convergent.
***
All the tests of convergence of infinite series discussed in Unit 8 for series of
positive terms can be used to decide absolute convergence of general series.

∞
cos nx
Example 5: Test the series ∑
n =1 n2
, x ∈ R, for convergence.

cos nx
Solution: Here an =
n2
∞
1 1
Since | cos nx | ≤ 1 , we get that | an | ≤
n2
. But ∑n
n =1
2
converges, (Why ?)
∞
This means that the series ∑a
n =1
n is absolutely convergent. Hence, the series
∞ ∞
cos nx
∑ an , i.e.,
n =1
∑
n =1 n2
is convergent. [See Theorem 2].

***
Now try the following exerices.

∞
1 1
E3) Test the conditional convergence of the series ∑ (−1)
n =1
n −1
 3 + 5 .
n n 
270
Unit 9 Alternating Series
n +1
(−1)
E4) Test the absolute convergence of the series ∑ 2n
.

E5) Determine the values of p for which the series

1 1 1 1
1− p
+ p − p + p L converges conditionally.
2 3 4 5

In the next section we shall see that we can generate new series from an
infinite series with entirely different nature.

9.4 REARRANGEMENT OF SERIES

In this section we shall discuss another aspect of an alternating series. This
involves rearrangement of the terms of the series. We shall explain this
through an example.

Consider the following series

1 1 1
1− + − +L ... (4)
2 3 4

Suppose we now rearrange the terms of the series by

1 1 1 1 1
1+ − + + − + ... ... (5)
3 2 5 7 4

This series can be written in Σ - notation by

∞
 1 1 1 
∑  4n − 3 + 4n − 1 − 2
n =1
n 


Then the series in (8) is known as a rearrangement (changing the order of the
terms) of the series in (7). Another rearrangement can be obtained by
rearranging the terms of the series (7) so that each positive term is followed by
two negative terms which gives the series

1 1 1 1 1 1 1 1
1− − + − − + − − + ...
2 4 3 6 8 5 10 12

In this way we can have different permutation of the terms and get different
rearranged series. You may note that this is very peculiar about an alternating
series. This does not happen with positive term series as all the terms are of
positive sign only.

We formally make a definition now.

∞
Definition 4: Let ∑u
n =1
n be series. Let σ be a one-to-one function from N
∞ ∞
onto N . Then ∑ σ π(n) is said to be a rearrangment of
n =1
∑u
n =1
n .

1 1 1 1 1
For example 1 + − + + − + L is a rearrangement of the series
3 2 5 7 4 271
Block 3 Infinite Series
1 1 1 1
1− + − + L.
2 3 4 5

Next we shall check whether the rearrangement of a series will preserve the
convergence or divergence nature or it will alter the sum of the series.

For that we shall again look into the series in (4) given by
1 1 1
1− + − +L ... (6)
2 3 4

We have seen in Sec. 9.2 that this series converges by Leibniz test. Infact this
series converges to ln 2 . The proof of this is beyond the scope of this course.

Let us now consider the rearrangement of the series given in (5). That is
1 1 1 1 1
1+ − + + − + ... ... (7)
3 2 5 7 4

Let (sn ) n denote the sequence of the partial sum of the series

1 1 1
1− + − +L,
2 3 4

and (t n ) n denote the sequence of the partial sum of the rearrangement of the
series

1 1 1 1 1
1+ − + + − + L.
3 2 5 7 4

1 1 1
We set rn = 1 + + + ... + − ln . Then (rn ) is convergent. We leave this as
2 3 n
an exercise for you to try (see E8).

We now have
1 1 1 1
s2 n = 1 − + −L + −
2 3 2n − 1 2n
1 1 1 1 1 1 
= 1 + + + ... + −  + + ... + 
3 5 2n − 1  2 4 2n 
 1 1  1 1 1  1 1 1 
= 1 + + ... +  −  + + ... +  −  + + ... + 
 2 2n   2 4 2n   2 4 2n 
 1 1  1 1 1 
= 1 + + L +  − 2 + + L + 
 2 2n   2 4 2n 
 1 1   1 1
= 1 + + L +  − 1 + + L + 
 2 2n   2 n
= [r2 n + ln 2n] − [rn + ln n]
= [r2 n − rn ] + ln 2n − ln n
2n
= [r2 n − rn ] + ln
n
= [r2 n − rn ] + ln 2
272
Unit 9 Alternating Series
Since (rn ) n is convergent, therefore (rn ) n is a Cauchy Sequence.
Consequently, there exists m ∈ N such that | r2n − rn | < ε for n ≥ m where
ε > 0 is any number.

This implies that lim s2 n = ln 2 .

n→∞

(−1) n+1
Now we consider the sequence ( s2 n+1 ). Since → 0 as n → ∞, we have
n
s2 n+1 − log 2 ≤ s2 n − log 2 + a2 n + 1 ≤ ε for n ≥ m.

Thus every partial sum of odd terms is within ε of log 2.

We have thus shown that lim sn = ln 2.

For the sequence (t n ), we have

 1 1 1 1 1  1 1 1 
t3n = 1 + −  +  + +  + L +  + − 
 3 2 5 7 4  4n − 3 4n − 1 2n 
 1 1 1 1 1  1 1 1 
= 1 + −  +  + −  + L +  + − 
 3 2 5 7 4  4n − 3 4n − 1 2n 
 1 1 1 1 1 1
= 1 + + + + + L + + 
 2 3 4 5 4n − 1 4n 
1 1 1  1 1 1
− 1 + + L +  − 1 + + L + 
2 2 2n  2  2 n

Thus,
1 1
t3n = (r4 n + ln 4n) − (r2n + ln 2n) − (rn + ln n)
2 2
 1 1  3
=  r4 n − r2 n − rn  + ln 2
 2 2  2

Again since (rn ) is a Cauchy Sequence, therefore,

3
lim t3n = ln 2 .
n→∞ 2
1 1 1
Since t3n+1 = t3n + and t3n+2 = t3n + + .
4n + 1 4n + 1 4 n + 3

3
Therefore lim t n = ln 2 .
n→∞ 2

That means the series in (7) which is the rearrangement of the series in (6)
3
converges to a different limit ln 2.
2
This shows that the rearrangement of conditionaly convergent series may
change its sum or limit.

Another rearrangement gives the series (with sign changed) is 273

Block 3 Infinite Series
1 1 1 1 1 1 1
+ + + ... + n − 1 + 8 + ... + 16 − + ...
2 4 6 2 2 +2 2 3

We state here without proof that this series is divergent. The proof is so
cumbursome that it is beyond the level of this course.

Hence the rearrangement not only changes the sum of the series but may
change its nature also. Now we ask under what conditions may we rearrange
the terms of the series without altering its value? An answer to this is given by
Riemann Rearragement Theorem which says that if an infinite series is
conditionally convergent, then its terms can be rearranged so that the new
series converges or diverges. The proof of the theorem is not easy and
therefore omitted.

We state two theorems (without proof) which will indicate the effect on the
convergence of a series if we rearrange the terms of the series.
∞
Theorem 3: If ∑u
n =1
n is an absolutely convergent series converging to s , then
∞
every rearrangement of ∑u
n =1
n also converges to s .

Thus, the order in which the terms occur is immaterial in absolutely convergent
series.

Is this result true for a conditionally convergent series? Towards an answer to

this question, we state the following result without proving it.

∞
Theorem 4(Riemann Rearrangement Theorem): Let ∑u
n =1
n be a conditionally

convergent series. Given any α ∈ R, there is a rearrangement of the series

∞

∑u
n =1
n which converges to α .

Bernhard Riemann You may note that we have already explained how the nature of the series
(1826-1866) changes with rearrangement for the series.

1 1 1 1 1
1+ − + + − +L .
3 2 5 7 4

You should be able to do the following exercise now.

1 1
E6) Show that ( rn ) n ∈r N where rn = 1 + + ... + − ln is convergent,
2 n
∞
E7) Suppose ∑u
n =1
n is a series of positive terms diverging to + ∞ . Show that
∞
every rearrangement of ∑u
n =1
n also diverges to + ∞ .

Let us briefly recall what we have covered in this unit.

274
Unit 9 Alternating Series

9.5 SUMMARY
In this unit we have covered the following:

1) We have introduced a special class of infinite series known as

alternating series – a series whose terms are alternatively positive and
negative.

2) Stated and proved Leibniz test for checking the convergence of an

alternating series.

3) We have discussed the two notions of convergence of an alternating

series.

i) Absolute convergence
ii) Conditional convergence

4) We have explained how the rearrangement of the terms of an alternating

series can change the nature of the original series.

9.6 SOLUTIONS/ANSWERS
1
E1) Here an = . Also an > an+1 > 0 .
n!
1
Also lim an = lim → 0 as n → ∞.
n →∞ n→∞ n!

∴ Therefore all the condition of Leibniz test are satisfying.

∴ the series converges.

E2) Here p > 0. Then

1 1 1 1
p
> p > p > ..... and lim p = 0.
1 2 3 n → ∞ n

Hence the series converges by the Leibniz test.

1 1 1 1 1 1
E3) Since 3
+ 5 > 3
+ 5
, and lim  3 + 5  = 0.
n n (n + 1) (n + 1) 
n →∞ n n 

∞
1 1
Therefore by Leibniz test, ∑ (−1)
n =2
n −1
 3 + 5  converges.
n n 
∞
1 1 ∞ 1 ∞
1
But ∑  3 + 5  = ∑ 3 + ∑ 5 also converges.
n =1  n n  n=1 n n =1 n

∞
1 1
∑ (− 1)
n+1
This means that  3 + 5  is absolutely convergent, but not
n =2 n n 
conditionally convergent.
∞
1
E4) The series ∑2
n= 0
n
is a convergent geometric series, so the series
275
Block 3 Infinite Series
n +1
(−1)
∑ 2n
is absolutely convergent.

E5) We have shown in E2 that the series converges for p > 0. But it is
absolutely convergent only for p ≥ 2. Hence this is conditionally
convergent for p ≥ 2 .

E6) We shall show that (rn ) is convergent when

1 1 1
rn = 1 + + + ... + − ln
2 3 n
n
1 dx
=∑ ∫
k1 x

n n
1 dx
rn = ∑ −∫
n =1 k 1 x

k +1
1 dx 1
Observe that
k +1
≤ ∫k
≤ .
x k
1 n+1 dx n dx 
n + 1  1 x ∫1 x 
rn+1 − rn = −∫ − 
n +1
1 dx
=
n +1
− ∫
n
x
1 1
{rn }n ≥1 ≤ − ≤0
n +1 n +1

Thus (rn ) is decreasing.

n k +1 dx
Also rn = ∑ − ∫
k =1 k x
n
1 n−1 1 1
≥∑ −∑ = ≥0
n =1 k k =1 k n

rn ≥ 0 ∀ n
∞ ∞
E7) Suppose a rearrangement ∑u
n =1
π(n) of ∑u
n =1
n converges.
∞ ∞
Note that ∑ un is itself a rearrangement of
n =1
∑u
n =1
π ( n) .
∞
Hence, ∑u
n =1
n would converge, contradicting the hypothesis that
∞

∑u
n =1
n diverges.
∞
Hence, every rearrangement of ∑u
n =1
n also diverges.

276
Block 3 Miscellaneous Examples and Exercises

MISCELLANEOUS EXAMPLES AND EXERCISES

The examples and exercises given below cover the concepts and processes
you have studied in this block. Doing them will give you a better understanding
of the concepts concerned, as well as practice in solving such problems.

We shall first consider miscellaneous examples. You can solve the exercises
in a similar way. We advise you not to look at the solutions of the exercises
given at the end unless you have tried to solve them on your own.

Miscellaneous Examples
Example 1: Check whether the following statements are true or false. Give
reasons for your answers.

1 1 1
a) The sequence of the partial sums of the series 1 + + + ... + + ... is not
2! 3! n!
a bounded sequence.

1 1 1
b) The series 1 − + − + ... is convergent.
3 5 7

c) The series ∑a n is convergent if lim an = 0.

n→∞

1 1 1
Solution: a) Let s n = 1 + + + ... + . Then (sn ) is the sequence of partial
2! 3! n!
sums which is monotonically increasing and is of positive terms. Therefore it
cannot be bounded. Thus the statement is true. We get that the statement is
true.

b) The series is an alternating series and an+1 ≤ a n ∀ n and

1
lim an = → 0 as n → ∞. Hence by the Leibnitz test, the series is
n→ ∞ 2n + 1
convergent. Therefore the statement is true.

∞
1 1 1
c) Let us consider the series ∑ n. Then a
n =1
n =
n
→ 0 as n → ∞. But ∑ n is
divergent. Hence the statement is false.
***
1 1 1
Example 2: Show that the series 1 − 2
+ 2
− + ... is convergent.
3× 2 5×3 7 × 42
1 1 1 (−1) n−1
Solution: Let ∑ an = 1 − +
3 × 2 2 5 × 32
− ... . Here a n =
2n − 1
×
n2
. Then

the series ∑a n is a series is an alternate terms and each of the terms is

positive and monotonically decreasing and lim an = 0. Therefore by Liebniz
n →∞
rule, the given series is convergent.

Since ∑u n converges and bn is positive and monotonically decreasing,

therefore by Abel’s test, the given series converge.

***
277
Block 3 Infinite Series
Example 3: Using the comparison test or the limit comparison test determine
the convergence of the following series

∞
1 ∞
cos 2 (2n)
i) ∑
n =1 n + n
ii) ∑
n =1 n3

1 1
Solution: i) Let an = and bn = , for n = 1,2,..., so that
n+ n n

an n
lim = lim
n →∞ b n →∞ n + n
n

1
= lim
n →∞ 1
1+
n
= 1 ≠ 0.
∞
1
Since ∑ n is divergent, we deduce, from the Limit Comparison Test, that
n =1
∞
1
∑n+
n =1 n
is divergent.

2
ii) We use the Comparison Test. Since 0 ≤ cos (2n) ≤ 1, for n = 1,2,..., we
cos 2 (2n) 1
have 0 ≤ ≤ 3 , for n = 1,2,....
n3 n
∞
1
Since ∑ 3 is convergent, we deduce, from the Comparison Test, that
n =1 n
∞
cos 2 (2n)
∑
n =1 n3
is convergent.

***
Example 4: Use Ratio test to determine whether the following series are
convergent.

∞
n3 ∞
n 2 2n
i) ∑
n =1 n !
ii) ∑
n =1 n !

n3
Solution: i) Let an = , for n = 1,2,..., so that
n!

an+1  (n + 1) 3   n ! 
= × 
an  (n + 1)!   n 3 
(n + 1) 2
=
n3
n 3 + 2n + 1 1 2 1
= 3
= + 2 + 3.
n n n n
a ∞
n3
Also lim n+1 = 0. Hence it follows, from the Ratio Test that
n →∞ a
∑
n =1 n !
is
n
convergent.
278
Block 3 Miscellaneous Examples and Exercises

n 2 2n
ii) Let an = , for n = 1,2,..., so that
n!
an+1  (n + 1) 2 2 n+1   n ! 
=   ×  2 n 
an  (n + 1)!   n 2 
2(n + 1)
=
n2
1 1 
= 2 + 2 .
n n 

a n+1 ∞
n 2 2n
Since lim
n →∞ a
= 0, it follows, from the Ratio Test that ∑
n =1 n !
is convergent.
n
***
Example 5: Prove that the following series is convergent and calculate its sum
∞
 1 3 
∑  2 n
+ .
n(n + 1) 
n =1 
∞ ∞
1 1
Solution: We know that ∑
n =1 2
n
is convergent, with sum 1, and that ∑ n(n + 1)
n =1
is convergent, with sum 1.

∞
 1 3 
Hence, by the Sum Rule and the Multiple Rule the series ∑  2 n
+  is
n(n + 1) 
n =1 
convergent, with sum 1 + (3 × 1) = 4 .
***
∞
n
Example 6: Show that ∑ (−1)
n =1
n +1

3n + 2
diverges.

n
Solution: The series is an alternating series. Here an = (−1) n+1 .
3n + 2

n n 1
Therefore a n = . Also lim an = lim = .
3n + 2 n →∞ n→∞ 3n + 2 3

Thus lim a n is not equal 0, therefore the series is divergent.

n →∞

***
∞
Example 7: Give an example of a series ∑u
n =1
n of positive terms such that

u n+1
< 1 for each n, but the series diverges. Does this example contradict the
un
Ratio Test? Justify.
1 u n+1 n
Solution: Let ∑ u = ∑ n , then
n
un
=
n +1
< 1 ∀ n.

1
But the series ∑ n is divergent.
277
Block 3 Infinite Series
This does not contradict the Ratio Test because when we take the limits as
u n+1
n → ∞, we find that lim = 1.
n →∞ u
n
***
Example 8: Test the convergence of the series whose general term is
−n 3 / 2
 1 
1 +  .
 n

1
Solution: Let u n =
n 3/ 2
 1 
1 + 
 n
1/ n 1
Then lim (u n ) = lim n
n→ ∞ n →∞
 1 
1 + 
 n
1
= n
 1 
lim 1 + 
n →∞
 n
1
= < 1.
e

Hence the series converges.

***
(−1)n
Example 9: Determine whether or not ∑n 2n
diverges, converges

conditionally, or converges absolutely.

(−1) n n n
Solution: Let an = n
. Then a n = n
2 2

n +1
a n+1 n +1 1 n +1
= 2 =
an n 2 n
n
2
a n+1 1
∴ lim → <1
n →∞ an 2

∴ The series is absolutely convergent and therefore convergent. Alternatively,

the lebeniz rule can also be applied to show that the series is convergent.
***
Example 10: Determine the convergence and divergence of the following
∞
n 2 − 3n + 4
series ∑
n =1 5n 4 − n
.

n 2 − 3n + 4 1
Solution: Let an = 4
and bn = 2 . Then both an and bn are positive
5n − n n

280
Block 3 Miscellaneous Examples and Exercises

an n 2 − 3n + 4 n 2
and = ×
bn 5n 4 − n 1

n 4 − 3n 3 + 4n 2
=
5n 4 − n
1 − 3n −1 + 4n −2
=
5 − n −3
an 1 1 1
Then lim
n →∞ a
→ as n → ∞. Since ≠ 0, and the series
5 5
∑n 2
is convergent,
n
the given series is convergent.
***
Miscellaneous Exercises
n2 −1 n
E1) Find the values of x for which the series ∑ n 2 + 1x converges.
E2) Prove that the following series is convergent and calculate its sum
∞   3 n 2 
∑   −

n =1   4  n(n + 1) 
.

E3) Interpret the decimal 0.999.... as infinite series, and hence represent
them as a fractions.

E4) Prove that the following series are convergent:

∞
(−1) n+1 n ∞
cos n
i) ∑ 3
n =1 n + 1
ii) ∑
n =1 2
n

E5) Determine which of the following series are convergent:

∞
(−1) n+1 n2
i) ∑
n =1 n
ii) ∑ 2n 2 + 1
–x–

SOLUTIONS/ANSWERS
E1) Hint: Apply D’Alembert’s Ratio test. Then the series converges for x < 1
and diverges if x > 1.

n
3 ∞
3
E2) The series ∑   is a geometric series, with a = r = . Hence, it is
n =1  4  4
3
convergent, with sum 4 = 3.
3
1−
4

277
Block 3 Infinite Series
∞
1
The series ∑ n(n + 1) is convergent, with sum 1 (cf. Sub-section 3.1.3).
n =1

  3 n ∞
2 
Hence, by the combination Rules ∑    − is convergent,

n =1   4  n(n + 1) 
with sum 3 − (2 × 1) = 1.
E3) We interpret 0.999... as

9 9 9
1
+ 2 + 3 + ... .
10 10 10
9 1 1
This is a geometric series with a = and r = . Since < 1, this
10 10 10
9
a
series is convergent with sum = 10 = 1; hence 0.999... = 1.
1− r 1− 1
10
n +1
(−1) n
E4) i) Hint: an = , n = 1, 2...
n3 + 1

n
an = 3
, n = 1,2...
n +1

Apply leibniz test. Then the series is absolutely convergent. Hence

it is convergent.

cos n
ii) Hint: Here an = .
2n
1
an ≤ since cos n ≤ 1.
2n
The series is absolutely convergent and hence convergent.
E5) i) Hint: Apply Leibniz test.

n2
ii) Let an =
2n 2 + 1

1 1
lim a n = lim =
n→ ∞ n→∞ 1 2
2+
n2

∴ the series is divergent.

282