All revision notes topics

Locating roots and iterationEdexcel A-Level Maths: Revision notes

Section 1

Locating roots by sign change

If ff is continuous on [a,b][a,b] and f(a)f(a), f(b)f(b) have opposite signs, then f(x)=0f(x)=0 has at least one root in [a,b][a,b]. Example: f(x)=x3−3x−5f(x)=x^3-3x-5 has f(2)=−3f(2)=-3 and f(3)=13f(3)=13, so a root lies in [2,3][2,3]. Narrowing further, f(2.2)=−0.952f(2.2)=-0.952 and f(2.3)=0.267f(2.3)=0.267 show 2.2<α<2.32.2<\alpha<2.3. Always state: the function is continuous, the values (or signs) at both ends, and the conclusion.

Key termscontinuouschange of sign
Common mistake

Saying 'there is a root' without mentioning continuity and the sign of each value.

Section 2

When sign change fails

The method can fail in two ways.

  • An interval that is too large may contain an even number of roots, so the signs at the ends match. For g(x)=(x−2)(x−3)g(x)=(x-2)(x-3), g(1.5)=g(3.5)=0.75g(1.5)=g(3.5)=0.75, yet the interval contains two roots. A narrower interval such as [1.5,2.5][1.5,2.5] shows the sign change: g(2.5)=−0.25g(2.5)=-0.25.
  • A discontinuous function can change sign across an asymptote without a root. f(x)=1x−2f(x)=\frac{1}{x-2} has f(1)=−1f(1)=-1, f(3)=1f(3)=1, but is never zero. A root where the curve touches the axis without crossing also gives no sign change.
Key termsasymptoteeven number of roots
Exam tip

If the signs match but you suspect a root, try a narrower interval.

Section 3

Iteration x(n+1) = g(x(n))

To solve f(x)=0f(x)=0 approximately, rearrange it as x=g(x)x=g(x) and iterate xn+1=g(xn)x_{n+1}=g(x_n) from a starting value x0x_0. For x3−3x−5=0x^3-3x-5=0, use x=3x+53x=\sqrt[3]{3x+5}: x0=2x_0=2, x1=2.2240x_1=2.2240, x2=2.2684x_2=2.2684, x3=2.2770x_3=2.2770, x4=2.2786x_4=2.2786. The values settle at α=2.28\alpha=2.28 (2 d.p.). For ex=4x\mathrm{e}^x=4x, two rearrangements x=ln⁡(4x)x=\ln(4x) and x=ex4x=\frac{\mathrm{e}^x}{4} find different roots. Use the answer button on your calculator (Ans\text{Ans}) to repeat the step, and keep full calculator accuracy.

Key termsiterationrearrangement
Common mistake

Rounding at every step, which can change the third decimal place.

Section 4

Cobweb and staircase diagrams

Draw y=g(x)y=g(x) and y=xy=x. Start at x0x_0 on the xx-axis, go vertically to the curve, then horizontally to the line y=xy=x, and repeat.

  • If 0<g′(x)<10<g'(x)<1 near the root, the path is a staircase moving monotonically towards the root.
  • If −1<g′(x)<0-1<g'(x)<0, the path spirals in as a cobweb.
  • If ∣g′(x)∣>1|g'(x)|>1, the path moves away: the iteration diverges. Example: for g(x)=ex4g(x)=\frac{\mathrm{e}^x}{4}, g′(x)=ex4g'(x)=\frac{\mathrm{e}^x}{4}, which equals xx at a root, so it converges to α≈0.36\alpha\approx0.36 (g′<1g'<1) and cannot find β≈2.15\beta\approx2.15 (g′>1g'>1).
Key termscobweb diagramstaircase diagramconvergence
Exam tip

Convergence condition: ∣g′(x)∣<1|g'(x)|<1 near the root. Say so in words and compare the gradient with 1.

Section 5

Why numerical methods are needed

Many equations, such as ex=4x\mathrm{e}^x=4x or x3−3x−5=0x^3-3x-5=0, cannot be solved by algebra. Numerical methods give a root to any required accuracy. To show a root is correct to nn decimal places, find a sign change over an interval of ±5\pm5 in the next decimal place: f(2.145)<0f(2.145)<0 and f(2.155)>0f(2.155)>0 show β=2.15\beta=2.15 to 2 d.p. State the accuracy asked for and show the values you used.

Key termsnumerical method
Exam tip

To justify 'correct to 2 d.p.', use an interval that runs half a unit either side, e.g. [2.145,2.155][2.145,2.155].

That's the notes covered.

Carry on to the next subtopic.

Exam questions on Locating roots and iteration

  1. Let f(x)=x3−3x−5f(x)=x^3-3x-5, a continuous function. The equation f(x)=0f(x)=0 has a single real root α\alpha.
    Show that α\alpha lies in the interval [2.2, 2.3][2.2,\,2.3].2 marks
  2. The root α\alpha of x3−3x−5=0x^3-3x-5=0 is estimated using the iteration xn+1=3xn+53x_{n+1}=\sqrt[3]{3x_n+5} with x0=2x_0=2.
    Find x3x_3 and x4x_4, and hence write down the value of α\alpha correct to 2 decimal places.2 marks
  3. Let f(x)=1x−2f(x)=\frac{1}{x-2} for x≠2x\ne2, and g(x)=x2−5x+6g(x)=x^2-5x+6.
    Show that f(1)f(1) and f(3)f(3) have opposite signs, and explain why this does not show that f(x)=0f(x)=0 has a root in [1,3][1,3].3 marks
See the full worksheet

Written by the Exaim team, led by Shaun Daswani (Head of Upper Secondary, Improve ME Institute; MSc Financial Mathematics, Imperial College London; BSc, UCL) and Jason Daswani (operational lead, Improve ME Institute; LSE).