Fixed Point Iteration

Numerical Methods

Source inspiration: (Mathew 2000-2019).

Animations

Each animation below shows the cobweb (staircase) diagram for fixed point iteration. The diagram plots two curves — the identity line \(y = x\) and the iteration function \(y = g(x)\) — then traces the staircase path: from the current iterate \(x_n\) on the diagonal, vertical to the curve to get \(g(x_n) = x_{n+1}\), then horizontal back to the diagonal. Convergence occurs when \(|g'(x^*)| < 1\).

Julia source scripts that generated these animations are linked under each case.

Case 1 — Convergent staircase, \(g(x) = \sqrt{2x}\), \(x_0 = 0.1\)

Behavior: The curve \(g(x) = \sqrt{2x}\) intersects \(y = x\) at \(x^* = 2\). Since \(g'(2) = \tfrac{1}{2} < 1\), iteration converges from below in a monotone staircase.

Julia source

Cobweb diagram: convergent fixed point iteration with g(x)=sqrt(2x), starting below the fixed point at x₀ = 0.1

Case 2 — Convergent staircase, \(g(x) = \sqrt{2x}\), \(x_0 = 3.8\)

Behavior: Same function as Case 1 but starting above the fixed point. The staircase descends monotonically to \(x^* = 2\), confirming convergence from both sides.

Julia source

Cobweb diagram: convergent fixed point iteration with g(x)=sqrt(2x), starting above the fixed point at x₀ = 3.8

Case 3 — Oscillating convergent, \(g(x) = \tfrac{8}{x+2}\), \(x_0 = 1.0\)

Behavior: The curve \(g(x) = \tfrac{8}{x+2}\) is strictly decreasing, so \(g'(2) = -\tfrac{1}{2}\). Since \(|g'(2)| = \tfrac{1}{2} < 1\), iteration converges, but the staircase alternates sides of \(x^*\) on each step.

Julia source

Cobweb diagram: oscillating convergent fixed point iteration with g(x)=8/(x+2), starting at x₀ = 1.0

Case 4 — Divergent, \(g(x) = \tfrac{x^2}{4} + \tfrac{x}{2}\), \(x_0 = 1.5\)

Behavior: The curve \(g(x) = \tfrac{x^2}{4} + \tfrac{x}{2}\) is tangent to \(y = x\) at \(x^* = 2\) from below. Since \(g'(2) = \tfrac{3}{2} > 1\), the staircase walks away from the fixed point toward the left.

Julia source

Cobweb diagram: divergent fixed point iteration with g(x)=x²/4+x/2, starting below the fixed point at x₀ = 1.5

Case 5 — Divergent, \(g(x) = \tfrac{x^2}{4} + \tfrac{x}{2}\), \(x_0 = 2.5\)

Behavior: Same function as Case 4 but starting above the fixed point. The staircase moves monotonically rightward and diverges, showing instability from both sides.

Julia source

Cobweb diagram: divergent fixed point iteration with g(x)=x²/4+x/2, starting above the fixed point at x₀ = 2.5

Case 6 — Oscillating divergent, \(g(x) = -\tfrac{x^2}{4} - \tfrac{x}{2} + 4\), \(x_0 = 1.5\)

Behavior: The curve \(g(x) = -\tfrac{x^2}{4} - \tfrac{x}{2} + 4\) is steeply decreasing at \(x^* = 2\). Since \(g'(2) = -\tfrac{3}{2}\) and \(|g'(2)| = \tfrac{3}{2} > 1\), iteration diverges while oscillating — the staircase spirals outward from the fixed point.

Julia source

Cobweb diagram: oscillating divergent fixed point iteration with g(x)=-x²/4-x/2+4, starting at x₀ = 1.5

References

Mathew, John H. 2000-2019. Numerical Analysis - Numerical Methods Modules. https://web.archive.org/web/20190808102217/http://mathfaculty.fullerton.edu/mathews/n2003/NumericalUndergradMod.html.