next up previous
Next: The Method of False Up: Bisection Methods: Previous: The Error Criterion

Convergence

Suppose our initial interval size, i.e., x2-x1, is given by I0. Then after the 1st bisection the interval size I1 = I0/2, after the second bisection, I2 = I1/2 = I0/4 etc. Generalizing, after the nth bisection,

In = In-1 / 2 = I0 / (2n). (2)

So if the tolerance level is delta , then the number of bisections required to reduce the interval width to delta is given by

Ndelta > log2(I0 / delta) (3)

Or in other words, if we think of the current interval size as the error, the error in the nth step is proportional to the error in the (n-1)th step. This is characteristic of a linearly convergent algorithm. We say that a method is linearly convergent if the error in the nth step is proportional to the error in the (n-1)th step raised to the first power, quadratically convergent if the error in the nth step is proportional to the error in the (n-1)th step raised to the second power, quartically convergent if raised to the fourth power, and so forth.


next up previous
Next: The Method of False Position Up: Bisection Methods: Previous: The Error Criterion
Mark D Smith
1998-10-01