Skip to content

Series ​

Numerical series ​

Definition ​

Let an∈K (n∈N) be a sequence. The expression

a1+a2+…+an=∑n=1∞an

is called an infinite sum or an infinite numerical series. The numbers an are called the terms of the series. The sequence

Sn=a1+a2+…+an=∑k=1nak(n∈N)

is called the partial sum sequence of the series. Sn is the n-th partial sum.

Convergence of series ​

The series ∑n=1∞an is called convergent if it's partial sum sequence (Sn) is convergent.

If the partial sum sequence does not converge, the series is said to be divergent.

Sum of an infinite series ​

The sum of an infinite series is the limit of the partial sum sequence

∑n=1∞an=limn→∞Sn=limn→∞∑k=1nak

where Sn is the n-th partial sum of the series.

Theorem (sum of two convergent series) ​

If the series ∑n=1∞an and ∑n=1∞bn are convergent, then the series ∑n=1∞(an+bn) is also convergent and

∑n=1∞(an+bn)=∑n=1∞an+∑n=1∞bn

Theorem (convergent series with scalar multiplication) ​

If the series ∑n=1∞an is convergent and c∈R, then the series ∑n=1∞c⋅an is also convergent and

∑n=1∞c⋅an=c⋅∑n=1∞an

Geometric series ​

Definition ​

Let q∈K be a fixed number. The series

q+q2+q3+⋯=∑n=1∞qn

is called a geometric series (with base q or with quotient q).

Theorem (convergence of geometric series) ​

The geometric series ∑n=1∞qn is convergent if |q|<1 and divergent if |q|≥1.

If the geometric series is convergent, the sum of the series is

∑n=1∞qn=q1−q

Proof ​

Suppose that q=1. Then

Sn=∑k=1n1k=n

which is an obviously divergent sequence.

Suppose that q≠1. Using formula for the sum of the first n terms of geometric sequence, we have

Sn=∑k=1nqk=q⋅qn−1q−1(n∈N)

It follows from the theorem about the convergency of geometric sequence that if |q|≥1, then (Sn) is divergent. If |q|<1, then

∑n=1∞qn=limn→∞Sn=limn→∞q⋅qn−1q−1=q⋅0−1q−1=q1−q

Zero sequence test ​

Let an∈K (n∈N) and suppose that series ∑n=1∞an is convergent. Then

limn→∞an=0

The converse statement is not true! ​

As a counterexample, consider the harmonic series

∑n=1∞1n

The terms of this series obviously tend to 0, but the series is divergent.

Proof ​

Let A=∑n=1∞an and (Sn) be the partial sum sequence of the series. Obviously

Sn=Sn−1+an(n∈N,n≥2)

Moreover we have

limn→∞Sn=Aandlimn→∞Sn−1=A

Therefore

limn→∞an=limn→∞(Sn−Sn−1)=limn→∞Sn−limn→∞Sn−1=A−A=0

Cauchy's convergence test ​

Let an∈K (n∈N). Then the series ∑n=1∞an is convergent if and only if

∀ε>0 ∃N∈N ∀m≥n≥N:|an+an+1+…+am|<ε

The number N is called threshold index.

Proof ​

Denote by (Sn) the sequence of partial sums: Sn=a1+a2+…+an. Since for m≥n≥2 we have

an+an+1+…+am=Sm−Sn−1

then the statement is simple consequence of the application of Cauchy's convergence criterion for sequence (Sn).

Positive term series ​

Definition ​

Let an∈R (n∈N). The series ∑n=1∞an is called a positive term series if an≥0 (n∈N).

Theorem ​

A positive term series is convergent if and only if it's partial sum sequence is bounded above.

Property of partial sums ​

Sn+1=Sn+an+1≥Sn(n∈N)

(Sn) is monotonically increasing.

Comparison test ​

Theorem ​

Let an,bn∈R (n∈N) be two sequences such that 0≤an≤bn (n∈N). Then

Majorant criterion ​

If the series ∑n=1∞bn is convergent, then the series ∑n=1∞an is also convergent.

Minorant criterion ​

If the series ∑n=1∞an is divergent, then the series ∑n=1∞bn is also divergent.

Proof ​

Denote by (Sn) and by Tn the partial sums of the series ∑n=1∞an and ∑n=1∞bn respectively. By the assumptions of the theorem (Sn) and (Tn) are monotonically increasing, furthermore

Sn≤Tn(n∈N)

a) If ∑n=1∞bn<∞, then (Tn) is bounded above. Therefore (Sn) is also bounded above, which implies that ∑n=1∞an<∞.

b) If ∑n=1∞an=∞, then (Sn) is not bounded above. Therefore (Tn) is also not bounded above, which implies that ∑n=1∞bn=∞.

Hyperharmonic series ​

Definition ​

Let p>0 be a fixed real number. The positive term series

∑n=1∞1np

is called a hyperharmonic series or p-series

Theorem ​

Let p>0. The hyperharmonic series

∑n=1∞1np

is convergent if and only if p>1.

Alternating series ​

A real number series is called an alternating series if it contains alternately positive and negative terms.

Definition ​

Let a1≥a2≥a3≥…>0 be a monotonically decreasing sequence of positive numbers. Then the series

a1−a2+a3−a4+a5−…=∑n=1∞(−1)n−1⋅an

is called a series of Leibniz type.

Theorem (Leibniz criterion) ​

The series of Leibniz type

∑n=1∞(−1)n−1an

is convergent if and only if limn→∞an=0.

Absolute and conditional convergence of a series ​

Absolute convergence ​

Definition ​

Let an∈K (n∈N). The series ∑n=1∞an is called absolutely convergent if the series of absolute values

∑n=1∞|an|

is convergent.

Theorem ​

If a series is absolutely convergent, then it is convergent.

Conditional convergence ​

Definition ​

A series that is convergent but not absolutely convergent is called conditionally convergent.

Example ​

The alternating hyperharmonic series ∑n=1∞(−1)n−1np is absolutely convergent if p>1 and conditionally convergent if 0<p≤1.

Rearrangement of series ​

Theorem ​

Let an∈K (n∈N), and p:N→N be a bijection. If the series ∑n=1∞an is absolutely convergent, then the series

∑n=1∞ap(i)

is absolutely convergent and is called a rearrangement of the series ∑n=1∞an.

Root test ​

Let an∈K (n∈N) and suppose that the limit

L=limn→∞|an|n∈[0,+∞]

exists. Then

  1. If L<1, the series ∑n=1∞an is absolutely convergent.
  2. If L>1, the series ∑n=1∞an is divergent. Moreover, limn→∞|an|=+∞.

Proof ​

  1. Suppose that L<1. Let q∈R, 0≤L<q<1. Then q−L>0, therefore
∃N∈N ∀n≥N:L−(q−L)<|an|n<L+(q−L)

The right side of inequality implies that

0≤|an|n<q(n≥N)

Taking the n-th power of the inequality, we get

|an|<qn(n≥N)

By 0≤q<1, the geometric series ∑n=1∞qn is convergent, then by majorant criterion the series ∑n=1∞|an| is also convergent.

  1. Suppose that L>1. Let q∈R, 1<q<L. Then L−q>0, therefore
∃N∈N ∀n≥N:L−(L−q)<|an|n<L+(L−q)

The left side of inequality implies that

|an|n>q(n≥N)

Taking the n-th power of the inequality, we get

|an|>qn(n≥N)

This inequality shows that (|an|) is not a zero sequence, therefore, by zero sequence test, the series ∑n=1∞an is divergent. Moreover, limn→∞|an|=+∞.

Ratio test ​

Let an∈K (n∈N) and suppose that the limit

L=limn→∞|an+1an|∈[0,+∞]

exists. Then

  1. If L<1, the series ∑n=1∞an is absolutely convergent.
  2. If L>1, the series ∑n=1∞an is divergent. Moreover, limn→∞|an|=+∞.

Proof ​

  1. Suppose that L<1. Let q∈R, 0≤L<q<1. Then q−L>0, therefore
∃N∈N ∀n≥N:L−(q−L)<|an+1an|<L+(q−L)

The right side of inequality implies that

0≤|an+1an|<q(n=N,N+1,N+2,…)

Fix a natural number k≥N+1, and write the above inequality for n=N,N+1,…,k−1. Then multiply these k−N inequalities with each other. We obtain that

|aN+1aN|⋅|aN+2aN+1|⋅…⋅|akak−1|<q⋅q⋅…⋅q⏟k−N

that is

|aN+1⋅aN+2⋅…⋅akaN⋅aN+1⋅…⋅ak−1|<qk−N

By the telescoping property of the left side of the inequality, we get

|akaN|<qkqN

that is

|ak|<|aN|qN⋅qk(k≥N+1)

By 0≤q<1, the series

∑n=1∞|aN|qN⋅qk=|aN|qN⋅∑n=1∞qk

is convergent, then by majorant criterion the series ∑n=1∞|an| is also convergent.

  1. Suppose that L>1. Let q∈R, 1<q<L. Then L−q>0, therefore
∃N∈N ∀n≥N:L−(L−q)<|an+1an|<L+(L−q)

The left side of inequality implies that

|an+1an|>q(n=N,N+1,N+2,…)

As in previous part fix k≥N+1, and write the above inequality for n=N,N+1,…,k−1. Then multiply these k−N inequalities with each other. We obtain that

|aN+1aN|⋅|aN+2aN+1|⋅…⋅|akak−1|>q⋅q⋅…⋅q⏟k−N

that is

|aN+1⋅aN+2⋅…⋅akaN⋅aN+1⋅…⋅ak−1|>qk−N

By the telescoping property of the left side of the inequality, we get

|akaN|>qkqN

that is

|ak|>|aN|qN⋅qk(k≥N+1)

This inequality shows that (|an|) is not a zero sequence, therefore, by zero sequence test, the series ∑n=1∞an is divergent. Moreover, limn→∞|an|=+∞.

Cauchy’s Product of two series ​

Definition ​

Let an,bn∈K (n∈N). Then the series

∑n=0∞∑i,j∈N∪{0}i+j=nai⋅bj

is called Cauchy's product of the series ∑i=0∞ai and ∑j=0∞bj.

The above formula is often written shortly as

∑n=0∞∑i+j=nai⋅bj

Theorem ​

Let an,bn∈K (n∈N). If the series ∑n=0∞an and ∑n=0∞bn are absolutely convergent, then the Cauchy's product of the two series is also absolutely convergent.

Power series ​

Definition ​

Let an∈K (n∈N) be a sequence and let x0∈K. The series

∑n=0∞an⋅(x−x0)n

is called a power series. The number an are the coefficients of the power series, x0 is the center of the power series, and x is the variable of the power series.

Convergence set ​

The set of all x∈K for which the power series converges is called the convergence set of the power series.

Example ​

∑n=0∞xnn!

is a power series. Here an=1n! and x0=0.

Theorem (convergence set of power series) ​

Let ∑n=0∞an⋅(x−x0)n be a power series and denote by S it's convergence set. Suppose that the limit exists:

L=limn→∞|an|n∈[0,+∞]

Then

  1. If L=0, the power series is absolutely convergent for all x∈K. Thus S=K.
  2. If L=+∞
    • the power series is absolutely convergent at x=x0.
    • the power series is divergent for all x≠x0.
  3. If 0<L<+∞, then the power series is
    • absolutely convergent for all x∈K such that |x−x0|<1L.
    • divergent for all x∈K such that |x−x0|>1L.

Thus B(x0,1L)⊆S⊆B―(x0,1L).

Proof ​

We have remarked that the power series is absolutely convergent at x=x0, thus x0∈S. Suppose that x∈K∖{x0} and apply the root test

∑n=0∞an⋅(x−x0)n

First we compute the limit of the n-th root

limn→∞|an⋅(x−x0)n|n=|x−x0|⋅limn→∞|an|n==|x−x0|⋅L

Then we can distinguish three cases:

  1. Suppose that L=0. Then
|x−x0|⋅L=0<1

therefore by root test the power series is absolutely convergent for all x∈K.

  1. Suppose that L=+∞. Then
|x−x0|⋅L=+∞>1(x≠x0)

therefore by root test the power series is divergent for all x∈K∖{x0}.

  1. Suppose that 0<L<+∞. Then
|x−x0|⋅L<1⟺|x−x0|<1L

this implies the statement of the theorem.

Radius of convergence ​

Using foregoing notations, suppose that limit L exists. Then the radius of convergence is defined as following:

R={0if L=+∞1Lif 0<L<+∞+∞if L=0

Analytical functions ​

Definition ​

Suppose that the radius of convergence of a power series is positive. Then it's sum function is called analytical function.

Exponential function ​

The function exp:K→K, defined as

exp(x)=1+x+x22!+x33!+…=∑n=0∞xnn!(x∈K)

is called the exponential function.

Sine function ​

The function sin:K→K, defined as

sin(x)=x−x33!+x55!−…=∑n=0∞(−1)n⋅x2n+1(2n+1)!(x∈K)

is called the sine function.

Cosine function ​

The function cos:K→K, defined as

cos(x)=1−x22!+x44!−…=∑n=0∞(−1)n⋅x2n(2n)!(x∈K)

is called the cosine function.

Hyperbolic sine function ​

The function sinh:K→K, defined as

sinh(x)=x+x33!+x55!+…=∑n=0∞x2n+1(2n+1)!(x∈K)

is called the hyperbolic sine function.

Hyperbolic cosine function ​

The function cosh:K→K, defined as

cosh(x)=1+x22!+x44!+…=∑n=0∞x2n(2n)!(x∈K)

is called the hyperbolic cosine function.

Theorem (convergence of exponential function) ​

The exponential function exp(x) is absolutely convergent for all x∈K.

Proof ​

Investigating convergency by ratio test, if x=0, then the series is absolutely convergent. If x∈K∖{0}, then

|xn+1(n+1)!xnn!|=|xn+1⋅n!(n+1)!⋅xn|=|x|n+1→0(n→∞)

Since 0<1, the series is absolutely convergent.

Simple properties of the functions above ​

  • exp(0)=1, sin(0)=0, cos(0)=1, sinh(0)=0, cosh(0)=1
  • For all x∈K,
sin(−x)=−sin(x), cos(−x)=cos(x)sinh(−x)=−sinh(x), cosh(−x)=cosh(x)

Theorem (connection between exp and sinh, cosh functions) ​

For all x∈K,

exp(x)=cosh(x)+sinh(x)

Proof ​

exp⁡x=∑n=0∞xnn!=∑n=0∞x2n(2n)!+∑n=0∞x2n+1(2n+1)!=cosh(x)+sinh(x)

Consequences ​

exp⁡(−x)=cosh(−x)+sinh(−x)=cosh(x)−sinh(x)exp⁡(x)+exp⁡(−x)=2⋅cosh(x)exp⁡(x)−exp⁡(−x)=2⋅sinh(x)cosh⁡(x)=exp⁡(x)+exp⁡(−x)2sinh⁡(x)=exp⁡(x)−exp⁡(−x)2

Theorem (connection between exp and sin, cos functions) ​

Let i be the imaginary unit. For all x∈K,

exp⁡(ix)=cos⁡(x)+i⋅sin⁡(x)Euler’s identity

Proof ​

exp⁡(ix)=∑n=0∞(ix)nn!=∑n=0∞(ix)2n(2n)!+∑n=0∞(ix)2n+1(2n+1)!==∑n=0∞i2n⋅x2n(2n)!+∑n=0∞i2n+1⋅x2n+1(2n+1)!=∑n=0∞(i2)n⋅x2n(2n)!+∑n=0∞i⋅(i2)n⋅x2n+1(2n+1)!==∑n=0∞(−1)n⋅x2n(2n)!+i⋅∑n=0∞(−1)n⋅x2n+1(2n+1)!=cos⁡(x)+i⋅sin⁡(x)

Consequences ​

exp⁡(−ix)=cos⁡(−x)+i⋅sin⁡(−x)=cos⁡(x)−i⋅sin⁡(x)exp⁡(ix)+exp⁡(−ix)=2⋅cos⁡(x)exp⁡(ix)−exp⁡(−ix)=2i⋅sin⁡(x)cos⁡(x)=exp⁡(ix)+exp⁡(−ix)2sin⁡(x)=exp⁡(ix)−exp⁡(−ix)2i

Theorem (addition of exponential functions) ​

For all x,y∈K,

exp⁡(x+y)=exp⁡(x)⋅exp⁡(y)

Proof ​

Apply Cauchy's product for the two absolutely convergent power series exp⁡(x) and exp⁡(y), then use Binomial theorem:

exp⁡(x)⋅exp⁡(y)=(∑n=0∞xnn!)⋅(∑n=0∞ynn!)=∑n=0∞∑k=0nxkk!⋅yn−k(n−k)!==∑n=0∞1n!⋅∑k=0nn!k!⋅(n−k)!⋅xk⋅yn−k=∑n=0∞1n!∑k=0n(nk)⋅xk⋅yn−k==∑n=0∞1n!⋅(x+y)n=∑n=0∞(x+y)nn!=exp⁡(x+y)

Theorem (addition of sin,cos,cosh,sinh functions) ​

For all x,y∈K,

sin(x+y)=sin(x)⋅cos(y)+cos(x)⋅sin(y)cos(x+y)=cos(x)⋅cos(y)−sin(x)⋅sin(y)sinh(x+y)=sinh(x)⋅cosh(y)+cosh(x)⋅sinh(y)cosh(x+y)=cosh(x)⋅cosh(y)+sinh(x)⋅sinh(y)

Proof ​

We prove cos⁡(x+y)=cos⁡(x)⋅cos⁡(y)−sin⁡(x)⋅sin⁡(y), the other statements are proved similarly. First we express cos with exp then we apply the addition of exponential functions theorem, then using Euler's identity we get back to cos and sin functions:

cos⁡(x+y)=exp⁡(i⋅(x+y))+exp⁡(−i⋅(x+y))2==exp⁡(ix+iy)+exp⁡(−ix−iy)2==exp⁡(ix)⋅exp⁡(iy)+exp⁡(−ix)⋅exp⁡(−iy)2==(cos⁡x+isin⁡x)(cos⁡y+isin⁡y)+(cos⁡x−isin⁡x)(cos⁡y−isin⁡y)2

Completing operations in the numerator, we get

2cos⁡xcos⁡y−2sin⁡xsin⁡y2=cos⁡xcos⁡y−sin⁡xsin⁡y

Corollary for sin and cos ​

If we apply the above theorem for x=y, we get

sin(2x)=2⋅sin(x)⋅cos(x)cos(2x)=cos2(x)−sin2(x)

If we apply the above theorem for x=−y, we get

1=cos2(x)+sin2(x)

Corollary for sinh and cosh ​

If we apply the above theorem for x=y, we get

sinh(2x)=2⋅sinh(x)⋅cosh(x)cosh(2x)=cosh2(x)+sinh2(x)

If we apply the above theorem for x=−y, we get

1=cosh2(x)−sinh2(x)

Theorem (value of exp⁡(1)) ​

By the definition of the exponential function

exp(1)=∑n=0∞1n!=e

where e is the Euler's number.

Theorem (irrationality of e) ​

Denote by Sn the partial sum of the series ∑n=0∞1n!, that is

Sn=∑k=0n1k!

Then for all n∈N

0<e−Sn<1n!⋅n

Proof ​

Sn can be regarded as an infinite series for any fixed n:

Sn=∑k=0n1k!=∑k=0∞akwhere ak={1k!if k≤n0if k>n

Therefore we have

e−Sn=∑k=n+1∞1k!−∑k=0∞ak=∑k=n+1∞(1k!−ak)=∑k=n+1∞1k!

Thus e−Sn>0. We can write that:

0<e−Sn=∑k=n+1∞1k!=1(n+1)!+1(n+2)!+…==1n!⋅(1n+1+1(n+1)(n+2)+1(n+1)(n+2)(n+3)+…)<<1n!⋅(1n+1+1(n+1)2+1(n+1)3+…)⏟geometric series==1n!⋅1n+11−1n+1=1n!⋅1n+1−1=1n!⋅n