The Newton-Raphson Method1 IntroductionThe Newton-

Raphson method, or Newton Method, is a powerful techniquefor

solving equations numerically. Like so much of the di erential
calculus,it is based on the simple idea of linear approximation.
The Newton Method,properly used, usually homes in on a root
with devastating eciency.The essential part of these notes is
Section 2.1, where the basic formulais derived, Section 2.2, where
the procedure is interpreted geometrically,and|of course|Section 6,
where the problems are. Peripheral but perhapsinteresting is
Section 3, where the birth of the Newton Method is described.

2 Using Linear Approximations to Solve Equa-tionsLetf(x) be a

well-behaved function, and letrbe a root of the equationf(x) = 0.
We start with an estimatex0ofr.Fromx0, we produce
animproved|we hope|estimatex1.Fromx1, we produce a new
estimatex2.Fromx2, we produce a new estimatex3. We go on until
we are `closeenough' tor|or until it becomes clear that we are
getting nowhere.The above general style of proceeding is
callediterative.Of the many it-erative root- nding procedures, the
Newton-Raphson method, with its com-bination of simplicity and
power, is the most widely used. Section 2.4 de-scribes another
iterative root- nding procedure, theSecant Method.Comment.The
initial estimate is sometimes calledx1, but most mathe-maticians
prefer to start counting at 0.Sometimes the initial estimate is
called a \guess." The Newton Methodis usually very very good
ifx0is close tor,andcanbehorridifitisnot.The \guess"x0should be
chosen with care.

2.1 The Newton-Raphson IterationLetx0be a good estimate ofrand

letr=x0+h. Sincethetruerootisr,andh=r−x0,thenumberhmeasures how far the
estimatex0is from thetruth.Sincehis `small,' we can use the linear (tangent
line) approximation toconclude that0=f(r)=f(x0+h)f(x0)+hf0(x0);and therefore,
unlessf0(x0)iscloseto0,h−f(x0)f0(x0):It follows thatr=x0+hx0−f(x0)f0(x0):Our
new improved (?) estimatex1ofris therefore given byx1=x0−f(x0)f0(x0):The
next estimatex2is obtained fromx1in exactly the same way asx1wasobtained
fromx0:x2=x1−f(x1)f0(x1):Continue in this way. Ifxnis the current estimate,
then the next estimatexn+1