g = cos x from 1
x₁ = 0.5403, x₂ = 0.8576
Cobweb spiral.
The iteration converges to a fixed point (started close enough) if , and diverges if . The error is multiplied by about at each step.
This page is easier if you are comfortable with:
For , the fixed point is and : the iteration converges, with errors alternating in sign and shrinking by a factor of 0.67 per step. For , and the error halves (the contraction mapping theorem). For , the fixed point is 1 but , so any other start moves away from it.
By the mean value theorem, (the mean value theorem), which gives the rule. If the convergence is faster than linear: this is what happens in Newton's method, , where at a simple root (Newton's quadratic convergence, Newton–Raphson).
Staircase and cobweb diagrams show the behaviour graphically (cobweb and staircase diagrams). The method is described in fixed-point iteration, and the global theory in Banach's fixed point theorem. Derivatives of trigonometric functions are in the derivative of sine.
A fixed-point iteration converges when the slope of the iteration function at the fixed point is less than 1 in size.
Iterates spiral in to the crossing point.
These examples are worked by hand, step by step, so you can compare your own working.
Find at .
Find .
Find at the fixed point 1.
A few situations come up again and again. Knowing them saves time and helps you spot answers that cannot be right.
x₁ = 0.5403, x₂ = 0.8576
Cobweb spiral.
|g′(x*)| = sin(0.739085) ≈ 0.6736
Linear convergence factor.
x* ≈ 0.739085
Solves cos x = x.
converges if |g′(x*)| < 1
Otherwise diverges.
Fixed-point iteration xₙ₊₁ = g(xₙ) converges to a fixed point x* when |g′(x*)| < 1. The error shrinks by about |g′(x*)| each step (linear convergence).
Iterate g = cos x from x₀ = 1. Find x₂ to 4 decimal places.
Find x₃.
Find the convergence factor |g′(x*)| at x* = 0.739085.
g(x) = 2x has fixed point 0. Is it attracting?
check |g(x*)| < 1 ✗
check |g′(x*)| < 1 ✓
The derivative controls the rate.
x₁ is the answer ✗
iterate until stable ✓
Check successive values.
any rearrangement converges ✗
some diverge ✓
Test |g′| < 1.
cos in degrees ✗
radians ✓
Use radians.
Try each one on paper, then open the answer. Use the solver above to see the full working.
0.8776
0.6390
0.6736
0.5
3
0.4537
Unlimited fresh questions with instant marking. Choose a level, work it out on paper, then type your answer.
Solve: …
When |g′| < 1 near the fixed point.
Linearly, with factor |g′(x*)|.
About 0.739085.
The iteration diverges.
Newton’s method.
cobweb and staircase diagrams.
More in Numerical methods and optimisation and the formula sheets.