Skip to content

Sequences ​

Index sequence ​

Definition ​

The sequence nk∈N is called an index sequence if it is strictly monotone increasing, that is

∀n∈N:nk<nk+1

Example ​

The sequence nk=2k is an index sequence.

Subsequence ​

Definition ​

Let a:N→H be a sequence and (nk) be an index sequence. Then the sequence

ank∈H(k∈N)

is called a subsequence of (an) (composed with the index sequence (nk)).

Example ​

Let an=1n (n∈N) and nk=2k, then

ank=a2k=12k=(12)k(k∈N)

is a subsequence of (an).

Neighborhood (environment) ​

Let us discuss the real number sequence an=1n (n∈N). We feel intuitively that the terms of this sequence are arbitrarily near to the number 0 if the index n is great enough. We say that the numbers 1n approach 0 or converge to 0. This impression is the base of the concept of the convergency and of the limit.

To define exactly what "near to a number" means, we need the concept of neighborhood (or ball or environment).

Definition ​

Let a∈K and r>0. The neighborhood of a with radius r is the set

B(a,r)={x∈K:|x−a|<r}⊂K

Neighborhood in R ​

If K=R, then the neighborhood of B(a,r) is the open interval (a−r,a+r).

Neighborhood in C ​

If K=C, then the neighborhood of B(a,r) is the open circular disk with center a and radius r on the complex number plane.

Convergence of a sequence ​

Definition ​

The number sequence an∈K (n∈N) is named convergent if

Definition using neighborhoods ​

∃A∈K ∀ε>0 ∃N∈N ∀n≥N:an∈B(A,ε)

Definition using inequalities ​

∃A∈K ∀ε>0 ∃N∈N ∀n≥N:|an−A|<ε

A number sequence is named divergent if it is not convergent.

Limit of a convergent sequence ​

The number A is called the limit of the convergent sequence an and is denoted as

A=limn→∞an

We often say that an tends to A, or an tends to A if n tends to infinity.

Boundedness of a sequence ​

The boundedness of a sequence is defined as the boundedness of its range.

Definition ​

The sequence an∈K (n∈N) is called bounded if

∃M>0 ∀n∈N:|an|≤M

The number M is called a bound of the sequence.

A number sequence is called unbounded if it is not bounded.

Theorem (boundedness of a convergent sequence) ​

Every convergent number sequence is bounded.

Proof ​

Let an∈K (n∈N) be a convergent sequence and A=limn→∞an∈K. Apply the definition of convergency with ε=1:

∃N∈N ∀n≥N:|an−A|<1

Using second triangle inequality, we get

|an|−|A|≤||an|−|A||<|an−A|<1

Rearrange:

|an|<1+|A|(n∈N,n≥N)

Thus obviously

|an|≤M(n∈N)whereM=max{|a1|,|a2|,…,|aN−1|,1+|A|}

Bolzano-Weierstrass theorem ​

Every bounded number sequence contains a convergent subsequence.

Convergency and ordering ​

Theorem ​

Let an, bn∈R (n∈N) be convergent sequences. And suppose that

limn→∞an≤limn→∞bn

Then

∃N∈N ∀n≥N:an≤bn

Proof ​

Let A=limn→∞an and B=limn→∞bn.

It is assumed that A≤B.

Let ε=B−A2>0.

Then, by the definition of the limit

∃N1∈N ∀n≥N1:|an−A|<B−A2

and

∃N2∈N ∀n≥N2:|bn−B|<B−A2

Using the definition of the absolute value, for any n≥N=max(N1,N2) we have

A−B−A2<an<A+B−A2B−B−A2<bn<B+B−A2

Since A+B−A2=B−B−A2=A+B2, we get

an<A+B2<bn(n∈N,n≥N)

Zero sequences ​

Definition ​

The sequence an∈K (n∈N) is called a zero sequence if it's convergent and limn→∞an=0.

Theorem (T1) ​

Let an∈K (n∈N) be sequence and A∈K. Then

limn→∞an=A⇔limn→∞(an−A)=0

Proof ​

The statement is a direct consequence of the definition of the limit and of the obvious identity

|an−A|=|(an−A)−0|

Theorem (T2) ​

Let an∈K (n∈N) be sequence. Then

limn→∞an=0⇔limn→∞|an|=0

Proof ​

The statement is a direct consequence of the definition of the limit and of the obvious identity

|an|=||an|−0|

Theorem (T3, Majorant Principle) ​

Let an∈K (n∈N) be sequence and bn∈K (n∈N) be a zero sequence. If

∃N0∈N ∀n≥N0:|an|≤bn

then (an) is also a zero sequence.

Proof ​

Let ε>0. Since limn→∞bn=0, then

∃N1∈N ∀n≥N1:|bn|<ε

Thus for the threshold index N=max(N0,N1) we have

|an−0|=|an|≤bn<ε

This means that limn→∞an=0.

Theorem (T4, Sum) ​

Let an,bn∈K (n∈N) be zero sequences. Then their sum (an+bn) is also a zero sequence.

Proof ​

Let ε>0. Since limn→∞an=0 then

∃N1∈N ∀n≥N1:|an|=|an−0|<ε2

and since limn→∞bn=0 then

∃N2∈N ∀n≥N2:|bn|=|bn−0|<ε2

Let N=max(N1,N2). It will be a good threshhold index, because, using the first triangle inequality, for any n≥N we have

|(an+bn)−0|=|an+bn|≤|an|+|bn|<ε2+ε2=ε

This means that limn→∞(an+bn)=0.

Theorem (T5, Product) ​

Let an∈K (n∈N) be a zero sequence and bn∈K (n∈N) be a bounded sequence. Then their product (an⋅bn) is also a zero sequence.

Proof ​

Let ε>0. Since (bn) is bounded, then

∃M>0 ∀n∈N:|bn|≤M

Since limn→∞an=0 then

∃N∈N ∀n≥N:|an|<εM

This N will be a good threshold index, because for any n≥N we have

|(an⋅bn)−0|=|an|⋅|bn|≤|an|⋅M<εM⋅M=ε

This means that limn→∞(an⋅bn)=0.

Operations with convergent sequences ​

Theorem (absolute value) ​

Let an∈K (n∈N) be a convergent sequence. Then the absolute value sequence (|an|) is also convergent and

limn→∞|an|=|limn→∞an|

Theorem (addition) ​

Let an,bn∈K (n∈N) be convergent sequences. Then their sum (an+bn) is also convergent and

limn→∞(an+bn)=limn→∞an+limn→∞bn

Proof ​

Let A=limn→∞an and B=limn→∞bn.

We have to prove that limn→∞(an+bn)=A+B.

Since, by T1, (an−A) and (bn−B) are zero sequences, then, by T4, the sequence

(an−A)+(bn−B)=(an+bn)−(A+B)

is also a zero sequence. By T1 we have

limn→∞(an+bn)=A+B

Theorem (multiplication) ​

Let an,bn∈K (n∈N) be convergent sequences. Then their product (an⋅bn) is also convergent and

limn→∞(an⋅bn)=limn→∞an⋅limn→∞bn

Proof ​

Let A=limn→∞an and B=limn→∞bn.

We have to prove that limn→∞(an⋅bn)=A⋅B.

Performing the following transformations

an⋅bn−A⋅B=an⋅bn−A⋅bn+A⋅bn−A⋅B=(an−A)⋅bn+A⋅(bn−B)

(an−A) and (bn−B) are zero sequences by T1.

The sequences (bn) and (A) are convergent, consequently, they are bounded. Thus, by T5, the sequences ((an−A)⋅bn) and (A⋅(bn−B)) are zero sequences.

Using T4, we get that (an⋅bn−A⋅B) is also a zero sequence.

Finally, by T1 we have

limn→∞(an⋅bn)=A⋅B

Theorem (division) ​

Let an,bn∈K (n∈N) be convergent sequences and bn∈K∖{0} (n∈N). Then their quotient (anbn) is also convergent and

limn→∞anbn=limn→∞anlimn→∞bn

Theorem (reciprocal) ​

Let bn∈K∖{0} (n∈N) be a convergent sequence. Suppose that B=limn→∞bn≠0. Then

  • The sequence (1bn) is bounded.
  • The sequence (1bn) is convergent and
limn→∞1bn=1B

Theorem (q-th root) ​

Let an∈R0+ (n∈N) be a convergent sequence and q∈N,q≥2. Then its q-th root sequence (anq) is also convergent and

limn→∞anq=limn→∞anq

Sandwich theorem ​

Let an,bn,cn∈K (n∈N) be sequences. Suppose that

∃N0∈N ∀n≥N0:an≤bn≤cn

and that (an) and (cn) are convergent and

limn→∞an=limn→∞cn=A

Then (bn) is also convergent and

limn→∞bn=A

Proof ​

Let's start with inequalities

an≤bn≤cn(n∈N,n≥N0)

After substracting an from all parts of the inequality, we get

0≤bn−an≤cn−an(n∈N,n≥N0)

Since

limn→∞(cn−an)=limn→∞cn−limn→∞an=A−A=0

then (cn−an) is a zero sequence. Using T3 we obtain that (bn−an) is also a zero sequence. Finally

limn→∞bn=limn→∞((bn−an)+an)=limn→∞(bn−an)+limn→∞an=0+A=A

Geometric sequence ​

Definition ​

Let q∈K be a constant. The sequence

an=qn(n∈N)

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

Theorem (convergence of geometric sequence) ​

The geometric sequence an=qn (n∈N) converges if and only if |q|<1 or q=1. In this case

limn→∞qn={0if |q|<11if q=1

Proof ​

The statement of the theorem is trivial if q=0 or q=1. Suppose that 0<|q|<1. Then 1|q|>1, and, using Bernoulli's inequality, we get

1|q|n=(1|q|)n=(1+1|q|−1)n≥1+n⋅(1|q|−1)>n⋅(1|q|−1)

After rearranging, we get

0≤|qn|=|q|n≤11|q|−1⋅1n(n∈N)

The right side sequence tends to 0. Using the sandwich theorem, we obtain limn→∞|qn|=0. Using T2, we get limn→∞qn=0.

Suppose that |q|>1. Once more using Bernoulli's inequality, we get

|qn|=|q|n=(1+|q|−1)n≥1+n⋅|q|>n⋅|q|

which implies that sequence (qn) is unbounded. Consequently, it is divergent.

Finally, suppose that |q|=1, but q≠1. Suppose indirectly that (an=qn) is convergent and denote by A its limit. Then by absolute value theorem we have

|A|=|limn→∞qn|=limn→∞|qn|=limn→∞|q|n=limn→∞1n=1

which implies A≠0.

On the other hand limn→∞an+1=limn→∞an=A, therefore

0=A−A=limn→∞an+1−limn→∞an=limn→∞(an+1−an)==limn→∞(qn+1−qn)=limn→∞qn⋅(q−1)=(q−1)⋅limn→∞qn=(q−1)⋅A

We obtained that

0=(q−1)⋅A

which is a contradiction, because on the right side stands the product of two non-zero numbers.

Other important convergent sequences ​

Theorem (convergence of an) ​

Let a∈R, a>0 be fixed. Then

limn→∞an=1

Proof ​

First case, suppose a>1.

Let n≥2 and apply the inequality between the arithmetic and geometric means for the n pieces of non-all-equal positive numbers

a,1,1,…,1:1<an=(a⋅1⋅…⋅1)n<a+n−1n=1+a−1n

The sequence on the right obviously tends to 1. Using the sandwich theorem, we obtain that limn→∞an=1.

Second case, suppose 0<a<1. This can be transformed to the first case by taking the reciprocal 1a>1. Then

an=11an=11an→11=1

Finally, the case a=1 is trivial.

Theorem (convergence of nn) ​

The sequence (nn) is convergent and

limn→∞nn=1

Proof ​

Suppose that n≥2. Using the inequality between the arithmetic and geometric means for the n pieces of non-all-equal positive numbers

n,n,1,1,…,1:1<nn=(n⋅n⋅1⋅…⋅1)n<2n+n−2n=2n+1−2n

The sequence on the right obviously tends to 1. Using the sandwich theorem, we obtain that limn→∞nn=1

Theorem (convergence of nk⋅qn) ​

Let q∈R, |q|<1 and k∈N be fixed. Then

limn→∞nk⋅qn=0

Proof ​

For any n∈N we have

|nk⋅qn|=nk⋅|qn|=((nn)n)k⋅|q|n=(nkn⋅|q|)n

Let s∈R be a number such that |q|<s<1. Since limn→∞nkn⋅|q|=1⋅|q|=|q|, then by definition of the limit

∃N∈N ∀n≥N:nkn⋅|q|<s

This means that for any n≥N we have

|nk⋅qn|=(nkn⋅|q|)n<sn

However, limn→∞sn=0, because (sn) is a geometric sequence with 0<s<1. Using the Majorant Principle, we obtain that (nk⋅qn) is a zero sequence.

Corollary (convergence of nkan) ​

Let a∈R, a>1. Applying the previous theorem with q=1a, we obtain for any k∈N that

limn→∞nkan=0

We can express this fact in this way: the exponential function (with base greater than 1) increases faster than a power function of any degree.

Theorem (convergence of xnn!) ​

Let x∈K be fixed. Then

limn→∞xnn!=0

Proof ​

Let n∈N and n≥|x|. (There exists such n by the Archimedean property of ordering.) Then we have for n≥N+2:

|xnn!|=|x|nn!=|x|⋅…⋅|x|⏞Nfactors⋅|x|⋅…⋅|x|1⋅2⋅…⋅N⏟Nfactors⋅(N+1)⋅…⋅n==|x|NN!⋅|x|N+1⋅|x|N+2⋅…⋅|x|n≤|x|NN!⋅|x|n→0(n→∞)

In the above estimation we have used that

|x|N+1<1, …, |x|n−1<1,

and that the other factors of the product are non-negative.

Finally, using the Majorant Principle, we obtain that (xnn!) is a zero sequence.

Monotone sequences ​

Definition ​

Let an∈R (n∈N) be a real number sequence. We say that this sequence is

  • monotonically increasing if ∀n∈N:an≤an+1.
  • strictly monotonically increasing if ∀n∈N:an<an+1.
  • monotonically decreasing if ∀n∈N:an≥an+1.
  • strictly monotonically decreasing if ∀n∈N:an>an+1.
  • monotone if it is monotonically increasing or monotonically decreasing.
  • strictly monotone if it is strictly monotonically increasing or strictly monotonically decreasing.

Theorem (monotone subsequence) ​

Every real number sequence has a monotone subsequence.

Theorem (convergence of monotonically increasing sequence) ​

Let an∈R (n∈N) be a monotonically increasing sequence. It is convergent if and only if it is bounded. Moreover

limn→∞an=sup {an:n∈N}∈R

Proof ​

Suppose that (an) is convergent. Then by boundedness of a convergent sequence theorem it is bounded. Because (an) is monotonically increasing, it is bounded above. Therefore it has finite supremum.

A=sup {an:n∈N}∈R

Let ε>0. Then A−ε<A, thus A−ε is not an upper bound. Therefore

∃N∈N:aN>A−ε

Since (an) is monotonically increasing, we obtain that ∀n≥N:an≥aN. Hence

A−ε<aN≤an≤A≤A+ε(∀n≥N)

which imples that |an−A|<ε. This means that, by definition of the limit,

limn→∞an=A

Theorem (convergence of monotonically decreasing sequence) ​

Let an∈R (n∈N) be a monotonically decreasing sequence. It is convergent if and only if it is bounded. Moreover

limn→∞an=inf {an:n∈N}∈R

Euler's number ​

Theorem ​

The sequence

an=(1+1n)n(n∈N)

is convergent.

Proof ​

We want to apply the monotone convergence theorem, therefore we will prove that (an) is monotonically increasing and bounded.

To prove that (an) is monotonically increasing, we will use the inequality between the arithmetic and geometric means for the n+1 pieces of non-all-equal positive numbers

1+1n,1+1n,…,1+1n,1

We obtain that

an=(1+1n)n==(1+1n)⋅…⋅(1+1n)⏟n times⋅1≤(n⋅(1+1n)+1n+1)n+1==(n+2n+1)n+1=(1+1n+1)n+1=an+1

This means that (an) is monotonically increasing.

To prove that (an) is bounded above, let us apply the inequality between the arithmetic and geometric means for the n+2 pieces of non-all-equal positive numbers

1+1n,1+1n,…,1+1n,12,12

We obtain that

14an=14(1+1n)n==(1+1n)⋅…⋅(1+1n)⏟n times⋅12⋅12≤(n⋅(1+1n)+1n+2)n+2==(n+2n+2)n+2=1

Hence we have an≤4 for any n∈N. This means that (an) is bounded above.

Finally, by the monotone convergence theorem, we obtain that (an) is convergent.

Definition ​

On the base of the previous theorem, we can define Euler's number e as

e=limn→∞(1+1n)n

Cauchy's sequence ​

Definition ​

The number sequence an∈K (n∈N) is called a Cauchy's sequence if

∀ε>0 ∃N∈N ∀m,n≥N:|an−am|<ε

Theorem ​

Every Cauchy sequence is bounded. Consequently, by the Bolzano-Weierstrass theorem, it has convergent subsequence.

Cauchy's convergence test ​

The number sequence an∈K (n∈N) is convergent if and only if it is a Cauchy's sequence.

Infinite limits of real number sequences ​

Definition ​

Definition for limn→∞an=+∞ ​

Let an∈R (n∈N) be a real number sequence. We say that (an) tends to +∞ if

∀P>0 ∃N∈N ∀n≥N:an>P

Definition for limn→∞an=−∞ ​

Let an∈R (n∈N) be a real number sequence. We say that (an) tends to −∞ if

∀P<0 ∃N∈N ∀n≥N:an<P

Theorem (limit of monotone increasing sequence) ​

Let an∈R (n∈N) be a monotonically increasing sequence. Suppose that it is not bounded above. Then

limn→∞an=+∞

Theorem (limit of monotone decreasing sequence) ​

Let an∈R (n∈N) be a monotonically decreasing sequence. Suppose that it is not bounded below. Then

limn→∞an=−∞

Example of non-monotone sequence with infinite limit ​

limn→∞n(−1)n=+∞wheren∈N

Theorem (addition) ​

Let an,bn∈R (n∈N) be sequences. Suppose that limn→∞an=A where −∞≤A≤+∞ and limn→∞bn=+∞. Then

limn→∞(an+bn)=+∞

Proof ​

Fix P1<A. Then by definition of the limit

∃N1∈N ∀n≥N1:an>P1

Let P>0. Since limn→∞bn=+∞, then to the real number P−P1 there corresponds a natural number N2∈N such that

∃N2∈N ∀n≥N2:bn>P−P1

Let N=max{N1,N2}. Then for any n≥N we have

an+bn>P1+P−P1=P

This means that limn→∞(an+bn)=+∞.

The table of addition in R― ​

x+yy=−∞y∈Ry=+∞
x=−∞−∞−∞not defined
x∈R−∞x+y+∞
x=+∞not defined+∞+∞

Sums (−∞)+(+∞) and (+∞)+(−∞) are called indeterminate sums. They are a kind of indeterminate expressions, since in that case sequence can behave itself variously.

For example

  • for an=n and bn=−nlimn→∞an=+∞,limn→∞bn=−∞an+bn=0limn→∞(an+bn)=0
  • for an=n and bn=−2nlimn→∞an=+∞,limn→∞bn=−∞an+bn=−nlimn→∞(an+bn)=−∞
  • for an=2n and bn=−nlimn→∞an=+∞,limn→∞bn=−∞an+bn=nlimn→∞(an+bn)=+∞

Theorem of addition for R― ​

If the real number sequences an,bn∈R (n∈N) have limits, furthermore the sum of limn→∞an and limn→∞bn is defined, then

limn→∞(an+bn)=limn→∞an+limn→∞bn

Theorem (multiplication) ​

Let an,bn∈R (n∈N) be sequences. Suppose that limn→∞an=A where A∈R―∖{0} and limn→∞bn=+∞. Then

limn→∞(an⋅bn)={+∞if A>0−∞if A<0