Newton's Method for Approximating the Root (Zero) of a Function

Caution: TEMATH's tools will write the values of computed results at the bottom of this Report window. This will cause you to constantly scroll between the computed results and the instructions for this activity. You may want to print a copy of the contents of the Report window before you begin this activity.

Newton's Method is defined by the iteration procedure

x[n+1] = x[n] - f(x[n])/f'(x[n]), n = 0, 1, 2, 3, ...

Using Newton's Method to Find the Square Root of a Number

The positive root of the function f(x) = x^2-2 is sqrt2. Let's use Newton's method to find this root.


How many iterations did it take to get 12 digit accuracy? Note that as the iterates get close to the root, the number of significant digits in the approximation doubles at each step.

Some of the Pitfalls of Newton's Method

1. Cycling

In this case, Newton's Method does not converge to the root using the initial approximation x0 = 0.


2. Dependency on the value of the initial approximation.

Note that it appears that Newton's Method is caught in a cycling mode. Be persistent and click at least 17 steps. Note that Newton's Method finally converges to the root at 0.

What happens if we change the value of the initial approximation by one unit in the fifth decimal place, that is, what will happen if we use the initial approximation x0 = 1.39175?

Note that it appears that Newton's Method is caught in a cycling mode. Be persistent and click at least 17 steps. Note that Newton's Method diverges.

3. Divergence

Note that in this case, no matter how close your initial approximation is to the root, Newton's Method will diverge. Why does this happen?
Return to Topic Start Return to Examples Menu

Copyright 2000-2008 Adam O. Hausknecht and Robert E. Kowalczyk