Relations

A relation can be said to generalize the concept of a function.

Definition

Definition

Let A be a set. A relation is a subset R of the cartesian product A \times A.

Any pair of elements a,b \in A is either belonging to R or not. If (a,b) \in R, we say “a is in the relation R to b”.

Notation

Let (a,b) belong to some relation R.

We notate either (a,b) \in R, or more frequently aRb.

Definition

Let (a_n) be a sequence of real numbers. We define a sequence (s_n) of partial sums by

s_n = \sum^n_{k=1} a_k = a_1 + a_2 + \cdots + a_n.

We refer to the sequence (s_n) as an infinite series.

We define (s_n) as a sequence of partial sums in order to formalize the series into a sequence, which we’ve already studied. All of our previous theorems will apply to (s_n), and therefore we can prove many theorems essentially for free!

Convergence

Definition

If (s_n) converges to a real number s, we say the series \sum a_n is convergent and write

\sum^\infty_{n=1}a_n = s.

If (s_n) does not converge, the series \sum^\infty_{n=1}a_n is divergent. If

\begin{align*} \lim_{n \to \infty} s_n &= +\infty \quad \text{then} \quad \sum^\infty_{n=1}a_n = +\infty \quad \text{ and}\\ \lim_{n \to \infty} s_n &= -\infty \quad \text{then} \quad \sum^\infty_{n=1}a_n = -\infty. \end{align*}

The question of whether or not a series converges is of great interest to mathematicians.

Example (Harmonic Series)

The harmonic series is defined

\sum^\infty_{n=1} \frac{1}{n}

The sequence (s_n) is divergent, and thus the harmonic series is divergent. This is

Example (Telescoping Series)

The series

\sum^\infty_{n=1} \frac{1}{n(n+1)}

has the partial sum

s_n = \frac{1}{2} + \frac{1}{6} + \frac{1}{12} + \cdots + \frac{1}{n(n+1)}

Theorem 5.1.1 (Algebraic properties of Series)

Let \sum a_n = s and \sum b_n = t. Then

(a) \sum (a_n + b_n) = s + t and

(b) \sum (ka_n) = ks for every k \in \mathbb{R}.

(a) By converting our series to limits, we obtain \sum (a_n + b_n) = \lim (s_{a_n} + s_{b_n}) = s + t.


(b) Similarly, we obtain

\sum (ka_n) = \lim ks_{a_n} = k\lim s_{a_n} = ks.

Theorem 5.1.2

If \sum a_n is a convergent series, then \lim_{n \to \infty} a_n = 0.

If \sum a_n is a convergent series, then the sequence of partial sums (s_n) must have a finite limit.

Since a_n = s_n - s_{n-1}, we have \lim a_n = \lim s_n - \lim s_{n-1} = 0.

Theorem 5.1.3 (Cauchy Criterion for Series)

The infinite series \sum a_n converges if and only if for each \varepsilon > 0, there exists a natural number N such that if n \geq m \geq N, then

|a_m + a_{m+1} + \cdots + a_{n}| = |s_n - s_{m-1}| < \varepsilon.

Let \sum a_n be a convergent series. Then the sequence (s_n) of partial sums converges. By the Cauchy Convergent Criterion, (s_n) is Cauchy. Thus, for any \varepsilon > 0, there exists N \in \mathbb{N} such that m,n \geq N implies |s_n - s_m| < \varepsilon. Hence, if n \geq m \geq N + 1, then m - 1 \geq N, and |s_n - s_{m-1}| < \varepsilon as desired.

Conversly, for all \varepsilon > 0 let N exists such that n \geq m \geq N implies |s_n - s_{m-1}| < \varepsilon. Then for n > m \geq N we have m + 1 > N, so that |s_n - s_m| < \varepsilon. This implies (s_n) is Cauchy, and therefore convergent.

Recap

Theorem 5.1.1 (Algebraic properties of Series)

Let \sum a_n = s and \sum b_n = t. Then

(a) \sum (a_n + b_n) = s + t and

(b) \sum (ka_n) = ks for every k \in \mathbb{R}.

Theorem 5.1.2

If \sum a_n is a convergent series, then \lim_{n \to \infty} a_n = 0.

Theorem 5.1.3 (Cauchy Criterion for Series)

The infinite series \sum a_n converges if and only if for each \varepsilon > 0, there exists a natural number N such that if n \geq m \geq N, then

|a_m + a_{m+1} + \cdots + a_{n}| = |s_n - s_{m-1}| < \varepsilon.