NUMERICAL SERIES

Theoretical notes Several ancient Greek philosophers concluded that perception is merely an appearance
and deceptive, as things are in constant changes, because by the time we perceive something, it has already
changed. Therefore, knowledge can only be acquired through thought. Zeno (c. 490–430 BCE), a representative
of the Eleatic school, became famous for formulating several paradoxes that point out the contradictions of
motion. In one of these, he claimed that a stone thrown at a tree can never reach its target, because it must
first cover half of the distance to the target, then half of the remaining distance, and so on. However, we cannot
add up infinitely many numbers.

...

0 s s s s s s
2 4 8 16 32

The distance covered approaches s

There is no doubt that a thrown stone can reach its target. So, what is wrong in the previous line of thought?
The motion was divided into infinitely many segments, and it was assumed that to complete the entire motion,
we would need to sum the lengths of each segment, which is impossible because there are infinitely many
segments. But in reality, no additions are needed to complete a motion; we do not stop along the way to
calculate. Additionally, he confused two types of infinity found in mathematics. One infinity pertains to the
number of segments in the journey, which relates to the cardinality of sets. The other infinity concerns the
unboundedness of the time required to complete the entire journey.
Zeno’s paradoxes draw our attention to the fact that nature can sum infinitely many numbers without performing
infinitely many calculations. It would be beneficial to develop a mathematical method capable of achieving this
as well. In this way, we could solve more practical problems with an entirely new approach. But first, we need
to define what we mean by the sum of infinitely many numbers. This brings us to the concept of a series.
A series is formed from a sequence whose terms we wish to ”add up”.
1. Given a sequence (an ) : N → R, the symbols
X X
an , or an , or a0 + a1 + a2 + · · · .
n=0

represent formally the sum of the terms of the sequence, which we call the infinite series (or simply
series) generated by (an ). On the other hand, the sequence

sn := a0 + a1 + a2 + · · · + an (n ∈ N)

is called the n-th partial sum of the series an , and an is the n-th term of the series an .
P P

2. We say that the series an is convergent if its partialP sums (sn ) is convergent, i.e., if ∃ lim(sn ) ∈ R.
P
In this case, we call this limit the sum of the series an , and denote it as follows:
∞
X
an := lim(sn ).
n=0

3. The series an is divergent if the sequence of partial sums (sn ) is divergent. In this case, the sequence
P
(sn ) either has no limit or
• lim(sn ) = +∞, in which case we say that the sum of the series an is +∞,
P

or
• lim(sn ) = −∞, in which case we say that the sum of the series an is −∞.
P

We denote these cases as

X∞ X∞
an := +∞, or an := −∞.
n=0 n=0

1
 4. Functions (an ) : {n ∈ Z | n ≥ M } → R are also considered sequences for any M ∈ Z. The sum of its
terms are represented formally by
X
an . or aM + aM +1 + aM +2 + · · ·
n=M

The sum of these series is defined in the same way as the limit lim(sn ), where
sn := aM + aM +1 + aM +2 + · · · + an (M ≤ n ∈ N),
is the n-th partial sum of the series. It is not hard to see the following transformation
∞
X ∞
X
an = an+M
n=M n=0

called reindexing. This means that the additional definitions and theorems for series started from n = 0
can be also applied (with appropriate modifications) for that series as well, though we will not emphasize
these separately.

Notable Series
q n , generated by the sequence (q n ), is called a
P
1. The Geometric Series. Let q ∈ R. The series
n=0
geometric series and is convergent if and only if |q| < 1, in which case the sum is
+∞
X 1
qn = 1 + q + q2 + q3 + · · · = |q| < 1 .


n=0
1−q
X ∞
X
If q ≥ 1, then the series q n diverges, and q n = +∞.
n=0 n=0
1
P
2. The Telescoping Series. The series n(n+1) , known as the telescoping series, is convergent, with a
n=1
sum of 1, that is,
∞
X 1 1 1 1
= + + + · · · = 1.
n=1
n · (n + 1) 1·2 2·3 3·4

3. The Hyperharmonic Series. Let α be a fixed real number. Then the series
X 1 1 1 1
= 1 + α + α + α + ···
n=1
n α 2 3 4

is called the hyperharmonic series and

∞
X 1
• diverges if α ≤ 1, in which case = +∞,
n=1
nα
• converges if α > 1.
Two special cases:
X1 1 1 1
• If α = 1, we have the harmonic series: = 1 + + + + · · · , which diverges.
n=1
n 2 3 4
X 1 1 1 1
• If α = 2, we have the superharmonic series: = 1 + 2 + 2 + 2 + · · · , which converges.
n=1
n 2 2 3 4
P 1
4. The Series Representation of e. The infinite series n! is convergent, and its sum is equal to the
n=0
n
number e := lim 1 + n1 , that is,


n→+∞

∞
X 1 1 1 1 1
= 1 + + + + + · · · = e.
n=0
n! 1! 2! 3! 4!

5. The Leibniz Series. The series

X (−1)n+1 1 1 1
= 1 − + − + ···
n=1
n 2 3 4

is known as the Leibniz series and is convergent.

2
Useful Results
1. Equiconvergence of Series. If two series differ by at most a finite number of terms, they are either both
convergent or both divergent.
P
2. Necessary Condition for Convergence of Series. If the infinite series an is convergent, then the
generating sequence (an ) is a null sequence, that is, lim(an ) = 0.
This statement immediately provides simple sufficient conditions P for divergence: If the sequence (an ) is not
a null sequence, or if it is divergent, then the infinite series an is divergent.
P P
3. Linear Combinations of Series. Assume that the sums of the series an and bn are
∞
X ∞
X
an =: A ∈ R és bn =: B ∈ R.
n=0 n=0

If λ, µ ∈ R are numbers such that λ · A + µ · B ∈ R is defined, then

+∞
X
λ an + µ bn = λ · A + µ · B.


n=0

4. Series with Non-negative Terms. A series with non-negative terms is convergent if and only if the
sequence of its partial sums is bounded.
P P
5. Comparison Test. Let an and bn be series with non-negative terms. Suppose that
∃N ∈ N, ∀n ≥ N : 0 ≤ an ≤ bn .
Then
P P
(a) Majorant Criterion: if the series bn is convergent, then an is also convergent.
P P
(b) Minorant Criterion: if the series an is divergent, then the series bn is also divergent.
6. Absolutely and Conditionally Convergent Series. We say that the series an is
P

• absolutely convergent if the series of absolute terms |an | is convergent,

P

• conditionally convergent if the series an is convergent but not absolutely convergent.

P

an is absolutely convergent, then it is also convergent. Note that Leibniz series is convergent,
P
If a series
but not absolutely convergent.
7. The Cauchy root test. Consider the infinite series an , and suppose that the limit
P

A := lim n |an | ∈ R
p
n→+∞

exists. Then
(a) If 0 ≤ A < 1, the series an is absolutely convergent (and therefore convergent as well).
P

(b) If A > 1, the series an is divergent.

P

(c) If A = 1, the series an may be either convergent or divergent.

P

8. The d’Alembert ratio test. Assume that in the series an , none of the terms are zero and that the limit
P

an+1
A := lim ∈R
n→+∞ an
exists. Then
(a) If 0 ≤ A < 1, the series an is absolutely convergent (and therefore convergent as well).
P

(b) If A > 1, the series an is divergent.

P

(c) If A = 1, the series an may be either convergent or divergent.

P

9. Leibniz Criterion. A series with alternating terms formed from a sequence that satisfies the condition
0 ≤ an+1 ≤ an (n ∈ N+ ) is called a Leibniz-type series:
X
(−1)n+1 an = a1 − a2 + a3 − a4 + · · ·
n=1

The Leibniz-type series (−1)n+1 an is convergent if and only if lim an = 0.

P
n=1 n→+∞

3
Exercise 1. Prove that the following infinite series are convergent, and evaluate their
sums:
 2
X (−5) n X (−1)n + 2n
a) , b) ,
n=2 3 5n+2
2n
n=0
X 1 X 1
c) , d) .
n=1 4n2 −1 n=1 n2 + 4n + 3

Solution

a) Transform the series as follows

(−5)n X (−5)n 5
 n
= =
X X
− .
n=2 3 9 9
2n n
n=2 n=2

P  n
This series is equiconvergent to − 59 because they differ only in the first two terms.
n=0
This last is a geometric series with ratio q = − 59 , which is convergent since |q| = 5
9
< 1.
Therefore, the given series is convergent, and its sum is:
∞
(−5)n ∞
5 ∞ 
5 ∞
5 5
 n n+2  2  n !
= = = =
X X X X
− − − · −
n=2 3 9 9 9 9
2n
n=2 n=0 n=0

5 ∞ 
5 25 1 25 9 25
 2 X n
= − · − = ·  = · = .
9 9 81 1 − − 81 14 126

5
n=0 9

b) First, we will transform the expression that defines the terms of the series as follows
 2
(−1)n + 2n 1 + 2 · (−2)n + 4n 1 1
 n
2 2
 n
1 4
 n
= = 2· + 2· − + 2· .
5n+2 5 ·5
2 n 5 5 5 5 5 5
This means that we can handle our series as the linear combination of geometric series
with ratios q = 15 , q = − 25 , and q = 45 . In each case, |q| < 1, so the series in question
are convergent, so the given series is also convergent, and its sum is
 2
+∞ (−1)n + 2n 1 +∞
X  1 n X  2 n
2 +∞ X  4 n
1 +∞
= 2· + 2· + 2· =
X
−
n=0 5n+2 5 n=0 5 5 n=0 5 5 n=0 5
1 1 2 1 1 1 1 2 1 43
= · + · + · = + + = .
5 1− 1
5 1 − −2 5 1− 4
20 35 5 140

2 2 2
5 5 5

c) We apply the method used for the telescoping series n(n+1) 1

, namely, we express the
P

n-th term of the series as the sum of two simpler fractions. We find A, B ∈ R such that
1 1 A B
= = + (n ∈ N).
4n2 −1 (2n − 1)(2n + 1) 2n − 1 2n + 1

4
 Determining A and B: After finding a common denominator,

1 A · (2n + 1) + B · (2n − 1) 2(A + B) · n + (A − B)

= = =⇒
(2n − 1)(2n + 1) (2n − 1)(2n + 1) (2n − 1)(2n + 1)
=⇒ 2(A + B) = 0 and A − B = 1 =⇒ A= 1
2
and B = − 12 , that is
1 1 1 1 1
= · − · .
4n2 −1 2 2n − 1 2 2n + 1
The n-th partial sum of the series 1
is therefore
P
4n2 −1

n n n
1 1 1 1 1 1X 1 1
   
sn = = = =
X X
· − · −
k=1
4k − 1 k=1 2 2k − 1 2 2k + 1
2 2 k=1 2k − 1 2k + 1

1 1 1 1 1 1 1 1 1 1 1
"      !#  
= −  +  −  +  −  + · · · + − = · 1− .
2 1 3 3 5 5 7 2n − 1

 2n + 1 2 2n + 1

Thus,
+∞
1 1 1 1
 
= lim sn = lim · 1− = .
X

n=1 4n − 1 2 2n + 1 2
2 n→+∞ n→+∞

d) Using the method above we obtain

1 1 A B A(n + 3) + B(n + 1)
= = + =
n2 + 4n + 3 (n + 1)(n + 3) n+1 n+3 (n + 1)(n + 3)
(A + B)n + 3A + B
= .
(n + 1)(n + 3)

Hence A + B = 0 and 3A + B = 1 from which we get A = 1/2 and B = −1/2. Hence

1 1/2 1 1 1
!
−1/2
 
= + =
X X X
−
n=1 n + 4n + 3 n=1 n + 1 n+3 n=1 2 n + 1 n+3
2

The n-th partial sum of this series is

n n
1 1 1 1X 1 1
   
sn = =
X
− −
k=1
2 k+1 k+3 2 k=1 k + 1 k + 3

1 1 1 1 1 1 1 1 1
       
= −  + −  +  −  +  −  + · · ·
2 2 4 3 5 4 6 5 7
1 1 1 1 1 1
!   !#
+ − +  − + −
n−1 n+1 n
 n + 2 n+1 n+3
1 1 1 1 1 1 1 1 5
   
= + − − −−−−−→ + −0−0 = .
2 2 3 n+2 n+3 n→+∞ 2 2 3 12

Therefore, the given series is convergent, and

∞
1 5
= .
X

n=1 n2 + 4n + 3 12

5
Exercise 2. The hour and minute hands of a traditional clock align at 12 o’clock. When
will the two hands align again?

Solution It takes 1 hour for the minute hand to return to the 12 o’clock position, by
which time the hour hand will be at the 1. After another 5 minutes ( 12
1
hour), the minute
hand will also reach the 1, but the hour hand will have moved further, and so on. Since
the minute hand moves 12 times faster than the hour hand, the minute hand will catch up
with the hour hand in
1 1 ∞
1 1 12
 n
1+ + 2 + ··· = = 1 = hours .
X
12 12 n=0 12 1 − 12 11

Exercise 3. Express the decimal x = 0.12̇4̇ as the ratio of two integers! (The dotted part
represents the repeating segment of the decimal).

Solution According to what we learned about geometric series,

1 1
 
0.12̇4̇ = 0.1 + 0.024 + 0.00024 + · · · = 0.1 + 0.024 1 + + + ...
100 1002
1 24 1 123
= + · 1 = .
10 1000 1 − 100 990

Exercise 4. Determine the convergence of the series below.

√ n
0.1,
X n
X
a) b) ,
n=1 n=1 2n + 1

1
X n+2
(−1) , n
1−
X
c) d) .
n=1 n=1 n

Solution We apply the necessary condition for the convergence of series. This states that
if the generating sequence (an ) is divergent or convergent but does not tend to zero, then
the series is divergent.

a) The general term of the series does not tend to zero, since
√
lim 0.1 = 1 ̸= 0.
n

n→+∞

Hence, the series is divergent.

b) The general term of the series does not tend to zero, since
n 1 1 1
lim = lim = = ̸= 0.
n→+∞ 2n + 1 n→+∞ 2 + 1
2+0 2
n

Hence, the series is divergent.

6
  
c) The general term (−1)n is divergent, hence the series is divergent.

d) The general term of the series does not tend to zero, since

1 1
n+2 " n  2 #
−1

lim 1− = lim 1+ · 1− = e−1 · (1 − 0)2 = e−1 ̸= 0.
n→+∞ n n→+∞ n n

Hence, the series is divergent.

Exercise 5. Determine the convergence of the series below.

X n2 + 2n − 5 X n2 + 2n − 5 X 1
a) , b) , c) ,
n=2 2n − n + 3 n=2 2n − n + 3
q
3 2 4 2
n=1 n(n + 3)
1 X √ √ 
n+1−
X
d) q , e) n .
n=3 n(n2 + 1) n=1

Solution We apply comparison test, meaning we find a convergent majorant or a divergent

minorant series for the given series. We say that bn is
P

• a majorant series of an if 0 ≤ an ≤ bn holds except for a finite number of

P

indices. In that case, if bn is convergent, then an is also convergent (majorant

P P

criterion).

• a minorant series of an if 0 ≤ bn ≤ an holds except for a finite number of

P

indices. In that case, if bn is divergent, then an is also divergent (minorant

P P

criterion).

To decide which criterion to apply, we must first estimate whether the series is convergent
or divergent, as this determines whether we should estimate the terms an from above or
from below. For this, we need to consider the dominant part of the expressions in an .

a) We guess that the series diverges, because

n2 + 2n − 5 n2 1 1
= if n is sufficiently large, and diverges.
X
≈ ,
2n − n + 3
3 2 2n 3 2n n=1 n

To prove that, we find a divergent minorant series. Since

n2 + 2n − 5 ≥ n2 + 0 = n2 and 2n3 − n2 + 3 ≤ 2n3 + 3 ≤ 2n3 + 3n3 = 5n3

(n≥3) (n≥1)

we have
n2 + 2n − 5 n2 1
an := ≥ = =: bn .
2n − n + 3 (n≥3) 5n
3 2 3 5n
According to the divergence of the harmonic series bn is divergent. Then, by mino-
P
n=1
rant criterion, an is also divergent.
P
n=2

7
b) We guess that the series converges, because

n2 + 2n − 5 n2 1 1
≈ 4 = 2, if n is sufficiently large, and converges.
X
2n − n + 3
4 2 2n 2n n=1 n
2

To prove that we find a convergent majorant series. Since

n2 + 2n − 5 ≤ n2 + 2n ≤ n2 + 2n2 = 3n2
(n≥1)

and
2n4 − n2 + 3 ≥ 2n4 − n2 = n4 + n4 − n2 = n4 + n2 (n2 − 1) ≥ n4 + 0 = n4
(n≥1)

we have
n2 + 2n − 5 3n2 3
an := ≤ = =: bn .
2n4 − n2 + 3 (n≥1) n4 n2
According to the convergence of the superharmonic series bn is convergent. Then,
P
n=1
by majorant criterion, an is also convergent.
P
n=2

c) We guess that the series diverges, because

1 1 1 1
= if n is sufficiently large, and diverges.
X
q ≈q ,
n(n + 3) n(n + 0) n n=1 n

To prove that we find a divergent minorant series:

1 1 1 1
an = q ≥ q =√ = =: bn .
n(n + 3) (n≥1) n(n + 3n) 4n2 2n

According to the divergence of the harmonic series bn is divergent. Then, by mino-

P
n=1
rant criterion, an is also divergent.
P
n=1

d) We guess that the series converges, because

1 1 1 1
= 3/2 , if n is sufficiently large, and converges.
X
q ≈q
n(n2 + 1) n(n2 + 0) n n=1 n3/2

To prove that we find a convergent majorant series:

1 1 1 1
an = q ≤ q = √ = 3/2 =: bn .
n(n2 + 1) (n≥1) n(n2 + 0) n3 n

According to the convergence of the hiperharmonic series for α = 3/2 we have that
bn is convergent. Then, by majorant criterion, an is also convergent.
P P
n=1 n=3

e) Using the transformation

√ √ n+1−n 1
n+1− n= √ √ =√ √
n+1+ n n+1+ n

8
 we guess that the series diverges, because
1 1 1 1
√ = 1/2 , if n is sufficiently large, and diverges.
X
√ √ ≈√
n+1+ n n+ n 2n n=1 n1/2

To prove that we find a divergent minorant series:

1 1 1 1
an = √ √ ≥ √ √ = √ = 1/2 =: bn .
n + 1 + n (n≥1) n + 3n + n 3 n 3n

According to the divergence of the hiperharmonic series for α = 1/2 we have that
P
bn
n=1
is divergent. Then, by minorant criterion, an is also divergent.
P
n=1
Remark It is also possible to determine the divergence of the series
X √ √ 
n+1− n
n=1

using the method applied for telescoping series. Indeed,

√ √  √ √  √ √  √ √ 
sn =  2 − 1 +  3 −  2 + · · · +  n − n−1 + n + 1 − 
n
√ √
= n + 1 − 1 −−−−→ +∞.
n→+∞

Exercise 6. Determine the convergence of the series below.

n2 n! X 1 1
n
+
X X
a) , b) , c) ,
n=0 2 n=1 100 2
n n n
n=1

(n!)2 (n!)2
n2 +n+1
X n
f)
X X
d) , e) , ,
n=2 (2n)! n=1 2 n+1
n2
n=0

X 4n · n! X 2n + 3n X 1−n
n
g) , h) , i) .
n=1 nn n=1 2n + 1 n=1 n2 + n

Solution We solve this exercise using ratio or root test.

a) Let us apply the Cauchy root test:

s √
n2 q n2 ( n n)2 12 1
an = n =⇒ |an | = = = = A.
n n
−−−−→
2 2n 2 n→+∞ 2 2
Since A < 1, the series is convergent.

b) Let us apply d’Alembert’s ratio test:

n! an+1 (n + 1)! 100n (n + 1) · n!  100

n
an = =⇒ = · = ·
100n an 100n+1 n! 100 · 
100
n
n!
n+1
= −−−−→ +∞ = A.
100 n→+∞
Since A > 1, the series is divergent.

9
c) Let us apply the Cauchy root test:
1 1 1 1 1 1
 n q
an = + =⇒ n
|an | = + −−−−→ + 0 = = A.
2 n 2 n n→+∞ 2 2
Since A < 1, the series is convergent.

d) Let us apply d’Alembert’s ratio test:

 2  2
(n!)2 an+1 (n + 1)! (2n)! (n + 1) · n! (2n)!
an = =⇒ =  · = ·
(2n)! an 2(n + 1) ! (n!)2 (2n + 2)! (n!)2

(n + 1)2 · 
(n!)
2
(2n)!
 (n + 1)
2
= ·  =
(2n + 2)(2n + 1) · (2n)!
  (n!)
2
2
(n+1)(2n

+ 1)
1+ 1
1+0 1
= n 
−−−−→ = = A.
2(2 + 0) 4

2 2 + n1 n→+∞

Since A < 1, the series is convergent.

e) Let us apply d’Alembert’s ratio test:

 2  2
(n!) 2
an+1 (n + 1)! 2 n2 (n + 1) · n! 2n
2

an = n2 =⇒ = · = ·
2 an 2(n+1)2 (n!)2 2n2 +2n+1 (n!)2

(n + 1)2 · 
(n!) 2n 1 n2 + 2n + 1
2 2

= · = ·

2n · 22n

2
·2 (n!)

 2
2 4n
1 2 1 n
  1
= n ( 4 ) + 2n( 14 )n + ( 14 )n −−−−→ (0 + 0 + 0) = 0 = A.
2 n→+∞ 2

Since A < 1, the series is convergent.

f) Let us apply the Cauchy root test:

n2 +n+1 n+1+1/n n s
n n n n n
 q  
an = =⇒ n
|an | = = · · n

n+1 n+1 n+1 n+1 n+1

1 1 1 1 1 1 1
= 1 n · · q −−−−→ · · = = A,
(1 + n ) 1 + 1
n
n
1+ 1 n→+∞ e 1 1 e
n

1 √
because xn := 1 + −→ 1 ∈ (0, +∞) =⇒ n xn −→ 1.
n
Since A < 1, the series is convergent.

g) Let us apply d’Alembert’s ratio test:

4n · n! an+1 4n+1 · (n + 1)! nn 4·

4n
(n+1)
· 
· n! nn
an = =⇒ = · = ·
nn an (n + 1)n+1 4n · n! (n+1)(n


+ 1)n 4

n
· n!
4nn 4 4 4
= =  n+1 n =  n −−−−→ = A.
(n + 1) n
1+ n1 n→+∞ e
n

Since A > 1, the series is divergent.

10
h) Let us apply the Cauchy root test:
v v
2 +3 2n+ 3n 3 +1
2n
3 (3) +1
s u 2 n
n n q u n
an = =⇒ |an | = = 3n
1 = √
n n
u
n
u
n
t · t
2n + 1 2n + 1 n 2+ n n
n 2 + n1
3
−−−−→ · 1 = 3 = A,
n→+∞ 1
( 23 )n + 1 0+1 1 √
because xn := −→ = ∈ (0, +∞) =⇒ n xn −→ 1.
2+ n 1
2+0 2
Since A > 1, the series is divergent.

i) Let us apply the Cauchy root test:

1−n 1−n n−1 n 1− 1

 n q
an = =⇒ n
|an | = 2 = 2 = 2· n
n2 + n n +n n +n n 1+ 1
n

1 1− 1
1−0
= · n
−−−−→ 0 · = 0 = A,
n 1+ 1
n
n→+∞ 1+0

Since A < 1, the series is convergent.

Exercise 7. Determine the convergence and the absolute convergence of the series below.

2n + 1 √ √  (−1)n · (n + 1)
(−1)n · n+1−
X X X
a) , b) n , c) .
n=1 (−3) n=1 2n + 5n + 2
n 2
n=1

Solution First we investigate the absolute convergence of the series, since the fact of
being absolutely convergent also implies that the series is convergent. Remember that the
series an is convergent, if |an | is convergent.
P P

If absolute convergence fails, we have to look for other methods to analyse the convergence
of the series. For instance, if we are dealing with a series with alternating terms, it may
be a good idea to consider the use of Leibniz Criterion. To do so, our series must have the
form
(−1)n+1 an = a1 − a2 + a3 − a4 + · · · ,
X

n=1

where (an ) is a positive and monotonically decreasing sequence. In this case the series is
convergent if and only if lim an = 0.
n→+∞

a) In this case we can apply the Cauchy root test to deal with the absolute convergence:
v s
2n + 1 2n + 1 2n + 1
q u
an = =⇒ |an | = =
n
u
n n
t
(−3)n (−3)n 3n
√ s
n
nn 1 1 1
= 2+ −−−−→ · 1 = = A,
3 n n→+∞ 3 3
1 √
because xn := 2 + −→ 2 + 0 = 2 ∈ (0, +∞) =⇒ n xn −→ 1.
n
Since A < 1, the series is absolute convergent, so it is also convergent.

11
b) We have already shown that the series
√ √  X √ √  X 1
(−1)n · n+1− n = n+1− n =
X
√ √
n=1 n=1 n=1 n+1+ n

is divergent. Therefore the given series is not absolute convergent. Coming back to
analysis of the original series, note that it may be written as
√ √  X 1
(−1)n · n+1− n = (−1)n · √
X
√
n=1 n=1 n+1+ n

which satisfies the condition of Leibniz Criterion. Indeed, it is obvious that the sequence
1
an = √ √ (n ∈ N+ )
n+1+ n

is positive, monotonically decreasing and

1 1 1 1
an = √ √ = 1/2 · q −−−−→ 0 · √ = 0.
n+1+ n n 1 + n1 + 1 n→+∞ 1+0+1

Therefore, the given series is convergent.

c) We guess that the series

(−1)n · (n + 1) n+1
=
X X

n=1 2n + 5n + 2
2
n=1 2n + 5n + 2
2

diverges, because
n+1 n 1 1
≈ 2 = if n is sufficiently large, and diverges.
X
,
2n2 + 5n + 2 2n 2n n=1 n

To prove that we find a divergent minorant series:

n+1 n n 1
|an | = ≥ = = =: bn .
2n2 + 5n + 2 (n≥1) 2n2 + 5n2 + 2n2 9n2 9n

According to the divergence of the harmonic series bn is divergent. Then, by mino-

P
n=1
rant criterion, |an | is also divergent, so the given series is not absolute convergent.
P
n=1
Coming back to analysis of the original series, note that it may be written as

(−1)n · (n + 1) X n+1
= (−1)n · 2
X

n=1 2n + 5n + 2 2n + 5n + 2
2
n=1

which satisfies the condition of Leibniz Criterion. It is obvious that the sequence
n+1
an = (n ∈ N+ )
2n2 + 5n + 2
is positive, but its monotonicity is not easy to determine at first sight. We prove that
(an ) is monotonically decreasing sequence, that is

an+1 − an ≤ 0 (n ∈ N+ ).

12
Indeed, for all n ∈ N+ we have

(n + 1) + 1 n+1
an+1 − an = − 2
2(n + 1) + 5(n + 1) + 2 2n + 5n + 2
2

n+2 n+1
= − 2
2n2 + 9n + 9 2n + 5n + 2
(n + 2)(2n2 + 5n + 2) − (n + 1)(2n2 + 9n + 9)
=
(2n2 + 9n + 9)(2n2 + 5n + 2)
2n2 + 6n + 5
=− < 0.
(2n2 + 9n + 9)(2n2 + 5n + 2)

On the other hand,

n+1 n 1 + n1 1 1 + n1 1+0
an = = · = · −−−−→ 0 · = 0.
2n2 + 5n + 2 n2 2 + n5 + 2
n2
n 2 + n5 + 2
n2
n→+∞ 2+0+0

Therefore, by Leibniz Criterion, the given series is convergent.

13