Integration

A partition $P$ of the interval $[a,b]$ is a set $\set{x_0, x_1, \dots, x_n}$ of real numbers such that

$x_0 < x_1 < \cdots < x_n, \quad x_0 = a, x_n = b.$

Let $f$ be a bounded function on $[a,b],$ and let $P = \set{x_0, x_1, \dots, x_n}$ be a partition of $[a,b].$ Then the upper sum (upper Darboux sum) of $f$ for the partition $P$ , $U(f,P),$ is given by

$U(f, P) = \sum_{k=1}^n \sup{\set{f(x) | x \in [x_{k-1}, x_k]}} (x_k - x_{k-1}),$

and the lower sum (lower Darboux sum) of $f$ for the partition $P$ $L(f,P)$ is given by

$L(f, P) = \sum_{k=1}^n \inf{\set{f(x) | x \in [x_{k-1}, x_k]}} (x_k - x_{k-1}).$

These sums essentially represent the area of rectangles bounded either above or below $f$ on each interval given by the partition.

Let $f$ be a bounded function on $[a,b],$ let $\mathcal{U}$ denote the set of upper sums for $f,$ and let $\mathcal{L}$ represent the set of lower sums for $f .$ Then the upper Darboux integral (or equivalently, upper Riemann integral) of $f$ on $[a,b]$ is given by

$\overline{\int_a^{b}} f(x) dx = \inf \mathcal{U},$

and the lower Darboux integral (or lower Riemann integral) of $f$ on $[a,b]$ is given by

$\underline{\int_a^{b}} f(x) dx = \sup \mathcal{L}.$

Theorem: For a given function an interval, the lower Riemann integral is always less than or equal to the upper Riemann integral:

$\underline{\int_a^{b}} f(x) dx \leq \overline{\int_a^{b}} f(x) dx.$

We say that $f$ is Riemann integrable on $[a,b]$ if

$\underline{\int_a^{b}} f(x) dx = \overline{\int_a^{b}} f(x) dx,$

and we say the Riemann integral of $f$ on $[a,b],$ is

$\int_a^{b} f(x) dx = \underline{\int_a^{b}} f(x) dx = \overline{\int_a^{b}} f(x) dx.$

Riemann's Condition Theorem: Let $f$ be a bounded function on $[a,b].$ Then $f$ is Riemann integrable on $[a,b]$ if and only if for every $\epsilon > 0,$ there exists a partition $P$ of $[a,b]$ such that

$U(f, P) - L(f, P) < \epsilon.$

Theorem: If $f$ is continuous on $[a,b],$ then $f$ is Riemann integrable on $[a,b].$

Riemann Integration Rules

Theorem: Assume $f$ and $g$ are Riemann integrable on $[a,b]$ and $c \in \reals.$ Then the following conditions hold

Linearity: $\int_a^{b} f(x) dx + c \int_a^{b} g(x) dx = \int_a^{b}(f(x) + cg(x))dx.$

$f \cdot g$ is Riemann integrable.

If $f(x) \leq g(x)$ for all $x \in [a,b],$ then $\int_a^{b} f(x) dx \leq \int_a^{b} g(x) dx.$

$\abs{f}$ is Riemann integrable and

$\abs{\int_a^{b} f(x) dx} \leq \int_a^{b} \abs{f(x)} dx.$

Integral Mean Value Theorem: If $f$ is continuous on $[a,b],$ there exists a number $c \in [a,b]$ such that

$\int_a^{b} f(x) dx = f(c)(b - a).$

Riemann Sums

Let $f$ be a bounded function on $[a,b]$ and $P = \set{x_0, x_1, \cdots, x_n}$ be a partition of $[a,b].$ Let $T = \set{t_1, t_2, \dots, t_n}$ where $t_k \in [x_{k-1}, x_k]$ for $k = 1,2,\dots, n.$ The sum

$\sum_{k=1}^n f(t_k)(x_{k-1} - x_k)$

is called the Riemann Sum for the partition $P$ and points $T,$ and is denoted by $S(f, P, T).$

Let $P = \set{x_0, x_1, \cdots, x_n}$ be a partition of $[a,b].$ The norm of $P$ is given by

$\norm{P} = \max{\set{\abs{x_k - x_{k-1}} | k \in \set{1,2,\dots,n}}}.$

The limit of the sum as the norm of the partition approaches zero is said to equal $I,$ written as

$\lim_{\norm{P} \to 0} S(f, P, T) = I,$

if for every $\epsilon > 0,$ there exists a $\delta > 0$ such that for any partition $P$ of $[a,b]$ with $\norm{P} < \delta$ and for any points $T,$ is is the case that

$\abs{S(f, P, T)} < \epsilon.$

Riemann Integral Evaluation Theorem: If $f$ is a Riemann integrable function on $[a,b],$ then

$\lim_{\norm{P} \to 0} S(f, P, T) = \int_a^{b} f(x) dx.$