Skip to content

Mathematics · Ch 9 — Applications of Integration

Riemann Integral

9.2.1

Riemann Integral

Setting up the Riemann sum. Let f(x)f(x) be a real-valued, bounded function on the closed interval [a,b][a,b], a<ba<b. Unlike in a first geometric picture, f(x)f(x) need not keep the same sign throughout [a,b][a,b] — it may take both positive and negative values.

Partition [a,b][a,b] into nn subintervals [x0,x1],[x1,x2],…,[xn−1,xn][x_0,x_1],[x_1,x_2],\ldots,[x_{n-1},x_n] with

a=x0<x1<x2<⋯<xn−1<xn=b.a=x_0<x_1<x_2<\cdots<x_{n-1}<x_n=b.

In each subinterval [xi−1,xi][x_{i-1},x_i], i=1,…,ni=1,\ldots,n, choose an arbitrary point ξi\xi_i with xi−1≤ξi≤xix_{i-1}\le\xi_i\le x_i, and form the sum

∑i=1nf(ξi)(xi−xi−1)=f(ξ1)(x1−x0)+f(ξ2)(x2−x1)+⋯+f(ξn)(xn−xn−1).…(1)\sum_{i=1}^n f(\xi_i)(x_i-x_{i-1}) = f(\xi_1)(x_1-x_0)+f(\xi_2)(x_2-x_1)+\cdots+f(\xi_n)(x_n-x_{n-1}). \quad\ldots(1)

This is called a Riemann sum of f(x)f(x) for the given partition. Because ξi\xi_i can be any point of [xi−1,xi][x_{i-1},x_i], there are infinitely many different Riemann sums for the same partition.

Definition of the Riemann (definite) integral. If, as n→∞n\to\infty with max⁡(xi−xi−1)→0\max(x_i-x_{i-1})\to0 (the widest subinterval shrinking to zero), the sum in (1) tends to a single finite value AA — regardless of how the partition and the points ξi\xi_i were chosen — then AA is called the definite integral of ff with respect to xx on [a,b][a,b], also called the Riemann integral, denoted

∫abf(x) dx,\int_a^b f(x)\,dx,

read "the integral of f(x)f(x) with respect to xx from aa to bb". When a=ba=b, ∫aaf(x) dx=0\displaystyle\int_a^a f(x)\,dx=0.

Note

This chapter works with f(x)f(x) continuous on [a,b][a,b]; the Riemann integral also exists more generally for bounded, piecewise-continuous ff. The integration variable is a dummy variable: ∫abf(x) dx=∫abf(u) du\int_a^b f(x)\,dx=\int_a^b f(u)\,du — it can be renamed freely.

Three standard choices of the evaluation point ξi\xi_i (all yield the same limit for continuous ff, but give different finite-nn approximations):

  • Left-end rule (ξi=xi−1\xi_i=x_{i-1}): ∫abf(x) dx=lim⁡n→∞∑i=1nf(xi−1)(xi−xi−1)\displaystyle\int_a^b f(x)\,dx=\lim_{n\to\infty}\sum_{i=1}^n f(x_{i-1})(x_i-x_{i-1}).
  • Right-end rule (ξi=xi\xi_i=x_i): ∫abf(x) dx=lim⁡n→∞∑i=1nf(xi)(xi−xi−1)\displaystyle\int_a^b f(x)\,dx=\lim_{n\to\infty}\sum_{i=1}^n f(x_i)(x_i-x_{i-1}).
  • Mid-point rule (ξi=12(xi−1+xi)\xi_i=\tfrac12(x_{i-1}+x_i)): ∫abf(x) dx=lim⁡n→∞∑i=1nf ⁣(xi−1+xi2)(xi−xi−1)\displaystyle\int_a^b f(x)\,dx=\lim_{n\to\infty}\sum_{i=1}^n f\!\left(\tfrac{x_{i-1}+x_i}2\right)(x_i-x_{i-1}).

Exercise 9.1 uses each rule without taking the limit — i.e. computes just one finite Riemann sum S=∑f(ξi)ΔxS=\sum f(\xi_i)\Delta x from a given 5-point partition of [1,1.5][1,1.5] — which is only an approximate value of the true integral (worked in the text as Example 9.1, estimating ∫00.5x2 dx\int_0^{0.5}x^2\,dx via all three rules on 5 equal subintervals of width h=0.1h=0.1).

Remarks connecting sign and area (used throughout §9.8):

  1. If ∫abf(x) dx\int_a^b f(x)\,dx exists, then F(x)=∫axf(u) duF(x)=\int_a^x f(u)\,du is a well-defined function on [a,b][a,b].
  2. If f(x)≥0f(x)\ge0 throughout [a,b][a,b], ∫abf(x) dx\int_a^b f(x)\,dx equals the geometric area between y=f(x)y=f(x), the xx-axis, and x=a, x=bx=a,\,x=b. …