Re: Secant method vs Newton's method

Schizoid Man wrote:
Hi,

Under what circumstances would I prefer the secant method to Newton's method?

Newton's method converges much faster and the number of function calls seem to be the same (if we assume the derivative is a function call).

When I can isolate a root I prefer a hybrid secant/binary search
method. The secant method can get "stuck" (of course it gets there
eventually) for certain functions, so putting in a binary search
step after each secant step speeds things up considerably in such
cases.

--
Julian V. Noble
Professor Emeritus of Physics
University of Virginia
.

Relevant Pages

  • Re: Secant method vs Newtons method
    ... Schizoid Man wrote: ... Under what circumstances would I prefer the secant method to Newton's ... shoots off to infinity. ...
    (sci.math.num-analysis)
  • Re: Secant method vs Newtons method
    ... Schizoid Man wrote: ... Under what circumstances would I prefer the secant method to Newton's ... trick (illinois) pretty fast, but unfortunately only 1D ...
    (sci.math.num-analysis)
  • Re: Secant method vs Newtons method
    ... In article, Schizoid Man ... Under what circumstances would I prefer the secant method to Newton's ... since you always have an upper and a lower bound for the zero. ... also entitled to reach the maximal accuracy on a given machine without ...
    (sci.math.num-analysis)
Loading