\( \newcommand{\matr}[1] {\mathbf{#1}} \newcommand{\vertbar} {\rule[-1ex]{0.5pt}{2.5ex}} \newcommand{\horzbar} {\rule[.5ex]{2.5ex}{0.5pt}} \newcommand{\E} {\mathrm{E}} \)
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::06. Limits of sequences

Square root, expressed as the limit of a sequence

Let \( c \in \mathbb{R} \) be a real. The sequence defined recursively below converges, and it converges to \( \sqrt{c} \).

Let \( x_1 = c \), and let \( x_n \) be defined like so:

\[ x_{n + 1} = \frac{1}{2}(x_n + \frac{c}{x_n}) \]

Proof by induction (any other typical methods) can be used to show that sequences defined recursively like this do (or don't) converge.

Intuition

The outer \( \frac{1}{2} \) factor can be seen as taking an average of the two inner terms. Inspecting the second term, \( \frac{1}{2}(c + \frac{c}{c}) = \frac{c + 1}{2} \), the third, \( \frac{1}{2}(\frac{c+1}{2} + c \frac{2}{c+1}) \), and so on, one can be convinced that this sequences is increasing starting from the third term. In addition, it can be seen that the terms don't surpass \( \sqrt{c} \). While not a exactly convincing as a proof, you can check if a term did surpass \( \sqrt{c} \), then the following term would be less than that term.


Source

Abbott, p54