forked from l0stman/sicp
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path2.13.tex
24 lines (22 loc) · 769 Bytes
/
2.13.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
\documentclass[a4paper,12pt]{article}
\usepackage{listings}
\lstset{language=Lisp}
\begin{document}
Suppose that all numbers are positive. Note $c_1$ and $p_1$ the
center and percentage tolerance of the interval $I_1$ and $c_2$ and
$p_2$ for the interval $I_2$.
We have,
\[ I_1 = [c_1(1-p_1), c_1(1+p_1)]\ \mbox{and}\
I_2 = [c_2(1-p_2), c_2(1+p_2)].\]
We then deduce the product $P$ of the two intervals
\[ P = [c_1c_2(1-p_1)(1-p_2), c_1c_2(1+p_1)(1+p_2)].\]
And finally the percentage tolerance $p$of $P$,
\begin{eqnarray*}
p &=& \frac{(1+p_1)(1+p_2) - (1-p_1)(1-p_2)}
{(1+p_1)(1+p_2) + (1-p_1)(1-p_2)} \\
&=& \frac{p_1+p_2}{1+p_1p_2}
\end{eqnarray*}
Under the assumption of small percentage tolerances, we then deduce
that
\[ p \simeq p_1 + p_2.\]
\end{document}