\( \newcommand{\matr}[1] {\mathbf{#1}} \newcommand{\vertbar} {\rule[-1ex]{0.5pt}{2.5ex}} \newcommand{\horzbar} {\rule[.5ex]{2.5ex}{0.5pt}} \newcommand{\E} {\mathrm{E}} \)
deepdream of
          a sidewalk
Show Question
\( \newcommand{\cat}[1] {\mathrm{#1}} \newcommand{\catobj}[1] {\operatorname{Obj}(\mathrm{#1})} \newcommand{\cathom}[1] {\operatorname{Hom}_{\cat{#1}}} \newcommand{\multiBetaReduction}[0] {\twoheadrightarrow_{\beta}} \newcommand{\betaReduction}[0] {\rightarrow_{\beta}} \newcommand{\betaEq}[0] {=_{\beta}} \newcommand{\string}[1] {\texttt{"}\mathtt{#1}\texttt{"}} \newcommand{\symbolq}[1] {\texttt{`}\mathtt{#1}\texttt{'}} \newcommand{\groupMul}[1] { \cdot_{\small{#1}}} \newcommand{\groupAdd}[1] { +_{\small{#1}}} \newcommand{\inv}[1] {#1^{-1} } \newcommand{\bm}[1] { \boldsymbol{#1} } \require{physics} \require{ams} \require{mathtools} \)
Math and science::Analysis::Tao::05. The real numbers

Cauchy sequences

A sequence \( (a_n)_{n=m}^{\infty} \) of rational numbers is a Cauchy Sequence iff the sequence is eventually ε-steady for every rational \( \varepsilon > 0 \). 

In other words, a sequence \( a_0, a_1, a_2, ... \) of rational numbers is a Cauchy Sequence iff for every rational \( \varepsilon > 0 \) there exists an integer N such that \( d(a_j - a_k) \le \varepsilon \) for every \( j, k \ge N \). Where \( d(a,b) = |a - b| \).

From sequences to reals

sequence → ε-steady sequence → eventually ε-steady sequence → Cauchy sequence → ε-close sequences → eventually ε-close sequences → equivalent sequences → real numbers.

Example

Proposition

The sequence \( a_1, a_2, a_3, ... \) defined by \( a_n := \frac{1}{n} \) (i.e., the sequence \( 1, \frac{1}{2}, \frac{1}{3}, ... \)) is a Cauchy sequence.

Proof
We have to show that for every \( \varepsilon > 0 \) there exists an integer \( N > 0 \) such that the sequence \( (a)_{N}^{\infty} \) is ε-steady, which would mean that \( d(a_j, a_k) \le \varepsilon \) for all \( j, k \ge N \):

\[ | \frac{1}{j} - \frac{1}{k}| \le \varepsilon \text{ for every } j, k \le N \]

As \( j, k \ge N \), we know that \( 0 \le \frac{1}{j}, \frac{1}{k} \le \frac{1}{N} \), so \( |\frac{1}{j} - \frac{1}{k}| \le \frac{1}{N} \). Thus, to force \( |\frac{1}{j} - \frac{1}{k}| \) to be less than or equal to ε, it is sufficient to choose an N such that \( \frac{1}{N} \) is less than ε, or in other words, an N such that \( N \gt \frac{1}{\varepsilon} \), and this can be done due to Prop. 4.4.1 (interspersing of integers by rationals). 


Source

Tao, Analysis I