UNIT 3 SOLUTION OF NON-LINEAR EQUATIONS

Structure Page Nos.

2.0 Introduction
2.1 Objectives
2.2 SOLUTION OF NONLINEAR EQUATIONS
2.3 Chord Methods For Finding Roots
2.3.1 Regula-falsi Method
2.3.2 Newton-Raphson Method
2.3.3 Secant Method
2.4 Iterative Methods and Convergence Criteria
2.4.1 Order of Convergence of Iterative Methods
2.4.2 Convergence of Fixed-point Method
2.4.3 Convergence of Newton’s Method
2.4.4 Rate of Convergence of Secant Method
2.5 Summary
2.6 Solutions/Answers

2.0 INTRODUCTION
In this unit we will discuss one of the most basic problems in numerical analysis. The problem is
called a root-finding problem and consists of finding values of the variable x (real) that satisfy the
equation f(x) = 0, for a given function f. Let f be a real-value function of a real variable. Any real
number  for which f() = 0 is called a root of that equation or a zero of f. We shall confine our
discussion to locating only the real roots of f(x), that is, locating non-real complex roots of f(x) = 0
will not be discussed. This is one of the oldest numerical approximation problems. The procedures
we will discuss range from the classical Newton-Raphson method developed primarily by Isaac
Newton over 300 years ago to methods that were established in the recent past.

Myriads of methods are available for locating zeros of functions and in first section we discuss
bisection methods and fixed point method. In the second section, Chord Method for finding roots will
be discussed. More specifically, we will take up regula-falsi method (or method of false position),
Newton-Raphson method, and secant method. In section 3, we will discuss error analysis for iterative
methods or convergence analysis of iterative method.

We shall consider the problem of numerical computation of the real roots of a given equation
f(x) = 0

which may be algebraic or transcendental. It will be assumed that the function f(x) is continuously
differentiable a sufficient number of times. Mostly, we shall confine to simple roots and indicate the
iteration function for multiple roots in case of Newton Raphson method.

All the methods for numerical solution of equations discussed here will consist of two steps. First step
is about the location of the roots, that is, rough approximate value of the roots are obtained as initial
approximation to a root. Second step consists of methods, which improve the rough value of each
root.
A method for improvement of the value of a root at a second step usually involves a
process of successive approximation of iteration. In such a process of successive approximation a
sequence {Xn} n = 0, 1, 2, … is generated by the method used starting with the initial approximation x o
of the root  obtained in the first step such that the sequence {X n} converges to  as n  . This xn is
called the nth approximation of nth iterate and it gives a sufficiently accurate value of the root .
For the first step we need the following theorem:

Theorem 1: If f(x) is continuous in the closed internal [a, b] and f(a) are of opposite signs, then there
is at least one real root  of the equation f(x) = 0 such that a <  < b.

If further f(x) is differentiable in the open interval (a, b) and either f’(x) < 0 or f’(x) > 0 in (a, b)
then f(x) is strictly monotonic in [a, b] and the root  is unique.

We shall not discuss the case of complex roots, roots of simultaneous equations nor shall we take up
cases when all roots are targeted at the same time, in this unit.

2.1 OBJECTIVES
After going through this unit, you should be able to:

 find an approximate real root of the equation f(x) = 0 by various methods;

 know the conditions under which the particular iterative process converges;
 define ‘order of convergence’ of an iterative method; and
know how fast an iterative method converges.

2.2: SOLUTION OF NONLINEAR EQUATIONS

UNIT 3 Let f(x) be a real-valued function of x defined over a finite interval. We assume it is
continuous and differentiable. If f(x) vanishes for some value x = , say, i.e. f() = 0, then we say x =
 is a root of the equation f(x) = 0 or that function f(x) has a zero at x = . We shall discuss methods
for finding the roots of an equation f(x) = 0 where f(x) may contain algebraic or transcendental
expressions. We shall be interested in real roots only. It is also assumed that the roots are simple (non-
repeated) and isolated and well-separated i.e. there is a finite neighbourhood about the root in which
no other root exists. All the methods discussed will be iterative type, i.e. we start from an approximate
value of the root and improve it by applying the method successively until two values agree within
desired accuracy. It is important to note that approximate root is not chosen arbitrarily. Instead, we
look for an interval in which only one root lies and choose the initial value suitably in that interval.
Usually we have to compute the function values at several points but sometimes we have to get the
approximate value graphically close to the exact root.
Method of Successive Substitution (Fixed Point Method)
Suppose we have to find the roots of the equation f(x) = 0. We express it in the form x =  (x) and the
iterative scheme is given as
x n  1   (x n )

where xn denotes the nth iterated value which is known and xn + 1 denotes (n + 1)th approximated value
which is to be computed. However, f(x) = 0 can be expressed in the form x =  (x) in many ways but
the corresponding iterative may not converge in all cases to the true value, rather it may diverge start
giving absurd values. It can be proved that necessary and sufficient condition for convergence of the
scheme is that the modulus of the first derivative of  (x) i.e.  (x) at the exact root should be less than
1 i.e. if  is the exact root then |  ()|  1 . But since we do not know the exact root which is to be
computed we test the condition for convergence at the initial approximation i.e. |  (x 0 )|  1 . Hence, it
is necessary that the initial approximation should be taken quite close to the exact root and test the
condition before starting the iteration. This method is also known as ‘fixed point’ method since the
mapping x   (x) maps the root  to itself since    ( ) i.e.  remains unchanged (fixed) under
the mapping x   (x) .
Example
Find the positive root of x 3  2x  8  0 by method of successive substitution correct upto two
places of decimal.
Solution
f (x)  x 3  2x  8

To find the approximate location of the root (+ ive) we try to evaluate the function
values at different x and tabulate as follows :
x 0 1 2 3 x>3
f(x) 8 9 4 13 + ive
Sign of f(x)    + +

The root lies between 2 and 3. Let us choose the initial approximation as x 0 = 2.5.
Let us express f(x) = 0 as x   (x) in the following forms and check whether |  ()|  1 for
x = 2.5.
(i) x  x3  x  8

1 3
(ii) x (x  8)
2
1
(iii) x  (2x  8) 3

We see that in cases (i) and (ii) |  (x)|  1 , hence we should discard these representations. As

1
the third case satisfies the condition, |  (x) |  2
 1 for x = 2.5 we have the
3(2x  8) 3

iteration scheme as,

1
x n  1  (2x n  8) 3

Starting from x0 = 2.5, we get the successive iterates as shown in the table below :
n 0 1 2 3
xn 2.5 2.35 2.33 2.33

Bisection Method (Method of Halving)

In this method we find an interval in which the root lies and that there is no other root in that interval.
Then we keep on narrowing down the interval to half at each successive iteration. We proceed as
follows :
(1) Find interval I  (x1 , x 2 ) in which the root of f(x) = 0 lies and that there is no other root
in I.
 x1  x 2
(2) Bisect the interval at x  and compute f(x). If | f(x) | is less than the desired
2
accuracy then it is the root of f(x) = 0.
(3) Otherwise check sign of f(x). If sign {f(x)} = sign {f(x 2)} then root lies in the interval
[x1, x] and if they are of opposite signs then the root lies in the interval [x, x 2]. Change
x to x2 or x1 accordingly. We may test sign of f(x)  f(x2) for same sign or opposite signs.
(4) Check the length of interval | x1 – x2 |. If an accuracy of say, two decimal places is
required then stop the process when the length of the interval is 0.005 or less. We may
x  x2
take the midvalue x  1 as the root of f(x) = 0. The convergence of this method is
2
very slow in the beginning.
Example
Find the positive root of the equation x 3  4x 2  10  0 by bisection method correct upto two
places of decimal.
Solution
f (x)  x 3  4x 2  10  0

Let us find location of the + ive roots.

x 0 1 2 >2
f(x)  10 5 14
Sign f(x)   + +
There is only one + ive root and it lies between 1 and 2. Let x 1 = 1 and x2 = 2; at x = 1, f(x) is
x  x2
– ive and at x = 2, f(x) is + ive. We examine the sign of f(x) at x  1  1.5 and check
2
whether the root lies in the interval (1, 1.5) or (1.5, 2). Let us show the computations in
the table below :
x1 +x2
Iteration No. x= Sign f(x) Sign f(x)  f(x2) x1 x2
2

1 1.5 + 2.375 + 1 1.5

2 1.25  1.797  1.25 1.5
3 1.375 + 0.162 + 1.25 1.375
4 1.3125  0.8484  1.3125 1.375
5 1.3438  0.3502  1.3438 1.375
6 1.3594  0.0960  1.3594 1.375
7 1.367  0.0471  1.367 1.375
8 1.371 + 0.0956 + 1.367 1.371

We see that | x1  x 2 |  0.004 .

1.367  1.371
We can choose the root as x   1.369 .
2
Regula-Falsi Method (or Method of False Position)
In this method also we find two values of x say x1 and x2 where function f(x) has opposite signs and
there is only one root in the interval (x1, x2). Let us express the function of y = f(x) and we are
interested in finding the value of x where curve y = f(x) intersects x-axis i.e. y = 0. We identify two
points (x1, y1) and (x2, y2) on the curve. Then we approximate the curve by a straight line joining these
two points. We find the point on the x-axis where this line cuts the x-axis. The equation of the straight
line passing through (x1, y1) and (x2, y2) is given by
y 2  y1
y  y1  (x  x1 )
x 2  x1
The point on x-axis where y = 0 is given by
x1 y 2  x 2 y1
x
y 2  y1
Now we check the sign of f(x) and proceed like in the bisection method. That is, if f(x) has same sign
as f(x2) then root lies in the interval (x1, x) and if they have opposite signs, then it lies in the interval
(x, x2). See Figure 1.
y

(1) y = f (x)
(2)
x1 x1 x1
x
O x2

Figure 1 : Regula-Falsi Method, Superscript Shows Iteration Number

Example
Find positive root of x 3  4x 2  10  0 by Regula-Falsi method. Compute upto the two
decimal places only.
Solution
It is the same problem as given in the previous example. We start by taking x 1 = 1 and x2 = 2.
We have y  x 3  4x 2  10 ; y1 =  5 and y2 = 14. The point on the curve are (1, – 5) and
(2, 14). The points on the x-axis where the line joining these two pints cuts it, is given by
I-Iteration
1  14  2  (  5) 24
x   1.26
14  5 19
y  f (x)   1.65
II-Iteration
Take points (1.26, – 1.65) and (2, 14)
 1.26  14  2  (  1.65)
x  1.34
14  1.65
y  f (x)   0.41
III-Iteration
Take two points (1.34, – 0.41) and (2, 14)
1.34  14  2  (  0.41)
x  1.36
14  0.41
y  f (x)   0.086
IV-Iteration
Take two points (1.36, – 0.086) and (2, 14)
1.36  14  2  ( 0.086)
x  1.36
14  0.086
Since value of x repeats we take the root as x = 1.36.
Secant Method
Like Regula-Falsi method in this method also two values of x, x 1 and x2 are chosen in the
neighbourhood of the actual root but they may be on the same side or on the opposite side of the
root. Then a straight line is drawn through (x1, y1) and (x2, y2) and position of x is found where it
intersects the x-axis. Then we take the points (x, y) and (x 1, y1) or (x2, y2) and draw straight line
and find point of intersection with x-axis and so on. See Figure 2.

y
(x1, y1) Secant

(x2, y2)

(1) (2) y = f (x)

x x
x
O

Figure 2 : Secant Method – Superscript Denotes Iteration Number

Example
Show four iterations of Secant method for finding the root of the equation x 3  4x 2  10  0
near x = 0 and x = 1. Compute upto two decimal places only.
Solution
f (x)  x 3  4x 2  10  0 ; y  f (x)  x 3  4x 2  10

y1  f (0)   10 ; y 2  f (1)   5
 We have two points on the curve (0, – 10) and (1, – 5) and can draw a secant passing through
these points. The point where it cuts x-axis is given by,
I-Iteration
0  ( 5)  1  ( 10)
x  2.0
 5  ( 10)
y  f (x)  14
II-Iteration
Take two points (1, – 5) and (2, 14)
1  14  2  ( 5)
x  1.26
14  5
y  f (x)   1.65
III-Iteration
Take two points (1, – 5) and (1.26, – 1.65)
1  ( 1.65)  1.26  (  5)
x  1.39
 1.65  5
y  f (x)  0.41
IV-Iteration
Take (1.26, – 1.65) and (1.39, 0.41)
1.26  0.41  1.39  ( 1.65)
x  1.36
0.41  1.65
y  f (x)   0.086
Newton-Raphson (N-R) Method
The Newton-Raphson’s method or commonly known as N-R method is most popular for finding
the roots of an equation. Its approach is different from all the methods discussed earlier in the
sense that it uses only one value of x in the neighbourhood of the root instead of two. We can
explain the method geometrically as follows :
Let us suppose we want to find out the root of an equation f(x) = 0 while y = f(x) represents a
curve and we are interested to find the point where it cuts the x-axis. Let x = x 0 be an initial
approximate value of the root close to the actual root. We evaluate y (x 0 )  f (x 0 )  y 0 (say).
dy
Then point (x0, y0) lies on the curve y = f(x). We find  f  (x) for x = x0, say f  (x 0 ) . Then
dx
we may draw a tangent at (x0, y0) given as,
y  y0  f  (x 0 ) (x  x 0 )
The point where the tangent cuts the x-axis (y = 0) is taken as the next estimate x = x 1 for the
root, i.e.
f (x 0 )
x1  x 0 
f  (x 0 )

f (x n )
In general x n  1  x n  , see Figure 3
f  (x n )
 y

y = f (x)

x1 x3 x4
x
O x2 x0

Figure 3 : Newton-Raphson Method

Theoretically, the N-R method may be explained as follows :

Let  be the exact root of f(x) = 0 and let  = x0 + h where h is a small number to be
determined. From Taylor’s series as have,
h2
f ()  f (x 0  h)  f (x 0 )  h f  (x 0 )  f  (x 0 )  . . .  0
2
f (x 0 )
Neglecting h2 and higher powers we get an approximate value of h, as h   . Hence, an
f  (x 0 )
approximation for the exact root  may be written as,
f (x 0 )
x1  x 0  h  x 0 
f  (x 0 )
In general the N-R formula may be written as,
f (x n )
xn  1  xn  , n  0, 1, 2, . . .
f  (x n )
It is same as derived above geometrically. It may be stated that the root of convergence of N-R
method is faster as compared to other methods. Further, comparing the N-R method with
method of successive substitution, it can be seen as iterative scheme for
f (x)
xx
f  (x)

f (x)
where  (x)  x 
f  (x)

The condition for convergence |  () |  1 in this case would be

{f  (x)}2  f (x) f  (x) f (x) f  (x)

 (x)  1   , at x = .
{f  (x)}2 {f  (x)}2

This implies that f  ()  0 .

Example
 Write N-R iterative scheme to find inverse of an integer number N. Hence, find inverse of 17
correct upto 4 places of decimal starting with 0.05.
Solution
Let inverse of N be x, so that we the equation to solve as,
x  N 1 or x 1  N  0
1
f (x)  x  1  N ; f  (x)  
x2
N-R scheme is
f (x n )  1 
xn  1  xn   x n  x 2n   N
f  (x n )  xn 
 x n (2  N x n )  x n (2  17x n )  N  17.
We take x0 = 0.05.
Substituting in the formula, we get
x1 = 0.0575 ; x2 = 0.0588 ; x3 = 0.0588
1
Hence,  0.0588 .
17
Example
Write down N-R iterative scheme for finding qth root of a positive number N. Hence, find
cuberoot of 10 correct upto 3 places of decimal taking initial estimate as 2.0.
Solution
1
We have to solve x  N q or x q  N  0

f (x)  x q  N ; f  (x)  q x q  1
The N-R iterative scheme may be written
x qn  N (q  1) x qn  N
xn  1  xn  
q x qn  1 q x qn  1
For cuberoot of 10 we have N = 10, q = 3.
2x 3n  10
Hence, xn  1 
3x 2n
Taking x0 = 2.0 we get the following iterated values
16 30.3520 29.9879
x1   2.167 ; x 2   2.154 ; x 3   2.154
2 14.0877 13.9191
1
Hence, we get 10 3  2.154 .

Example
Using N-R method find the root of the equation x – cos x = 0 correct upto two places of decimal

only. Take the starting value as ( = 3.1416,  radian = 180o).
4
Solution
 f (x)  x  cos x ; f  (x)  1  sin x
N-R scheme is given by
x n  cos x n x n sin x n  cos x n
xn  1  xn  
1  sin x n 1  sin x n


Taking x 0 
4
    1  1
sin  cos
x1  4 4 4  4 2 2
 1
1  sin 1
4 2

1
1.7854
 4   0.7395
2  1 2.4142

0.7395 sin (0.7395)  cos (0.7395)

x2 
1  sin (0.7395)

0.7395  0.6724  0.7449

  0.7427
1  0.6724
Upto two places of decimal the root is 0.74.
Note : If starting value is not given, we can plot graphs of y = x and y = cos x and locate their
point of intersection which will be root of x – cos x = 0. See Figure 4

y= x

y = cos x

x
O /2

Figure 4 : Intersection of y = x and y = cos x

Supplementary material

2.2 ITERATIVE METHODS FOR LOCATING ROOTS

One of the most frequently occurring problem in scientific work is to find the roots of equations of the
form:
f(x) = 0

In other words, we want to locate zeros of the function f(x). The function f(x) may be a polynomial in
x or a transcendental function. Rarely it may be possible to obtain the exact roots of f(x) = 0. In
general, we aim to obtain only approximate solutions using some computational techniques. However,
it should be borne in mind that the roots can be computed as close to the exact roots as we wish
through these methods. We say x* satisfies f(x) = 0 approximately when f(x*) is small or a point x*
which is close to a solution of f(x) = 0 in some sense like x* –  <  where  is a root of
f(x) = 0.

To find an initial approximation of the root, we use tabulation method or graphical method which
gives an interval containing the root. In this section, we discuss two iterative methods (i) bisection
method and (ii) fixed-point method. In a later section we shall discuss about the rate of convergence
of these methods.

2.2.1 Bisection Method

Suppose a continuous function f, defined on the interval [a, b], is given, with f(a) and f(b) of opposite
signs, i.e. f(a) f(b) < 0, then by Intermediate Value Theorem stated below, there exists a number  on
the real line such that a <  < b, for which f() = 0.
Theorem 2 (Intermediate-value Theorem): If the function f is continuous on the closed interval [a,
b], and if f(a)  y  f(a), then there exists a point c such that a  c  b and f(c) = y.

The method calls for a repeated halving of subintervals of [a, b] and, at each step, locating the “half”
1
containing . To start with, a1 = a and b1 = b, and let 1 be the mid point of [a, b], that is 1 = (a1 +
2
b1). If f(1) = 0, then  = 1. If not, then f(1) has the same sign as either f(a1) or f(b1). If f(a1)
f(1) < 0, then root lies in (a1, 1). Otherwise the root lies in (1, b1). In the first case we set a2 = a1
and b2 = 1 and in the later case we set a2 = 1 and b2 = b1. Now we reapply the process to the
interval (a2, b2). Repeat the procedure until the interval width is as small as we desire. At each step,
bisection halves the length of the preceding interval. After n steps, the original interval length will be
1
reduced by a factor enclosing the root.
2n

f(b) (b,f(b))

u2 x
a u1 u3 b

f(a)
(a,f(a))

Figure 1: Bisection Method

We now mention some stopping procedures that could be applied to terminate the algorithm. Select a
tolerance  > 0 and generate 1, 2, … n until one of the following conditions is met:
(i) n – n–1 < , (2.2.1)
 α n  α n 1
(ii) < , n  0, or (2.2.2)
αn
(iii) f(n) < ε (2.2.3)

While applying bisection method we repeatedly apply a fixed sequence of steps. Such a method is
called an Iteration method.

However, it is pertinent to mention that difficulties can arise using any of these stopping criteria. For
example, there exist sequence {n} with the property that the differences n – n–1 converge to zero
while the sequence itself diverges. Also it is possible for f(n) to be close to zero while n differs
significantly from . The criteria given by (2.2.2) is the best stopping criterion to apply since it tests
relative error.

Though bisection algorithm is conceptually clear, it has significant drawbacks. It is very slow in
converging. But, the method will always converge to a solution and for this reason it is often used to
obtain a first approximation for more efficient methods that are going to the discussed.

Theorem 3: Let f  C [a, b] and suppose f(a).f(b) < 0. The bisection procedure generates a sequence
{n} approximating  with the property,

ba
n –   , n  1.
2n

Now we illustrate the procedure with the help of an example.

Example 1

Find the least positive root of equation.

f(x) = x3 + 4x2 – 10 = 0

Check that f(x) has only one root in the interval in which this least positive root lies. Find 4 by
bisection algorithm.

Solution

Consider the following tables of values.

X 0 1 2
f(x) –10 –5 14

Take a1 = 1, b1 = 2, since f(a1) f(b1) < 0.

We give the four iterations in the following tabular form.

N an bn n f(n)
1 1 2 1.5 2.375
2 1 1.5 1.25 – 1.79687
3 1.25 1.5 1.375 0.16211
 4 1.25 1.375 1.3125 – 0.84839
5 1.3125 1.375 1.34375 – 0.35098

After four iterations, we have 4 = 1.3125 approximating the root  with an error
 – 4 1.375 – 1.325 = .050 and since 1.3125 < .

α  α4 b5  a 5 .050 .5 1 –1 1 1–2
< <  = 10 = 10
α a5 1.3125 10 2 2

That is, the approximation is correct to at least 2 significant digits.

Remarks 1: Generally the first stage methods for location of the roots of f(x) = 0 are (i) Tabulation
method and (ii) Graphical method. The method of tabulation is very crude and labourious and we
have used it in the above example to some extent in locating the least positive root of f(x) = 0. In
graphical method we plot the graph of the curve y = f(x) on the graph paper and the points where the
curve crosses the x-axis gives approximate values of the roots.

2.2.2 Fixed-point Method (or Method of Interation)

This method is also known as Method of Successive Approximations or Method of Iteration. In this
method, we write the equation f(x) = 0.

For example x3 – x – 1 = 0 can be written as,

1
x = (1  x) 3
1 x
or x=
x2
1 x
or x=
x

Now solving f(x) = 0 is equivalent to solving x = g(x).

Each such g(x) given above is called an iteration function. In fact, these are infinite number of ways
in which the original equation f(x) = 0 can be written as x = g(x). Out of all these functions where one
is to be selected, will be discussed in the following analysis.

Definition 1: A number  is called a fixed point of g(x) if g() =  and g is called the iteration
function.

Our problem is now to find out fixed point(s) of g(x). Graphically x = g(x) is equivalent to solving y =
x and y = g(x).
 y

y=x

y = g(x)

a b x

Figure 2: Fixed Point Method

Once an iteration function is chosen, to solve x = g(x), we start with some suitable value x 0 close to the
root (how to choose this will be explained) and calculate
x1 = g(x0) (the first approximation), then x2 = g(x1) (second approximation) and so on.

In general

xn+1 = g(xn), n = 0, 1, 2 …

The sequence {xn} converges (under some suitable conditions on g) to a number  (say). If g is
continuous then this gives  = g(), that is,  is a fixed point of g(x).

Concerning the existence, uniqueness of a fixed point and convergence of the sequence, we state a
theorem below:

Theorem 4 (Fixed Point Theorem): Let iteration function g(x) be defined and continuous on a
closed interval I = [a, b]. suppose further that g(x) satisfies the following:
(i) g(x)  I for all  I
(ii) g(x) is differentiable on I = [a, b]
and there exists a non-negative number k < 1 such that for all x  I, g(x)  k < 1.

Then
(a) g(x) has a fixed point ,
(b) the fixed point is unique, and
(c) the sequence {x1} generated from the rule xn+1 = g(xn) converges to , the fixed point of g(x),
when x0  [a, b]

Proof: (a) Existence: Suppose g(a) = a or g(b) = b, then there is nothing to be proved. So, suppose
g(a)  a and g(b)  b. Then g(a) > a and g(b) < b since g(x)  I for all x  I.

Consider h(x) = g(x) – x

Then h(a) = g(a) – a > 0 and

h(b) = g(b) – b < 0

Also h(x) is continuous on I since g(x) is so. Hence by Intermediate Value Theorem, there exists a
number , a <  < b such that h() = 0 
 g() –  = 0, i.e., g() = 

Hence g(x) has a fixed point in I.

(b) Uniqueness: From (2.2.4)

h(x) = g(x) – 1, but g(x)  k < 1

Hence h(x) < 0.

Therefore, h(x) is a decreasing function and it crosses x-axis only once, i.e. h(x) vanishes only once in
I.

Therefore g(x) – x = 0 only for unique value of x in (a, b). Hence uniqueness.

(c) Convergence: Let  be the fixed point of g(x). We have

 = g() and xn+1 = g(xn).

Let en+1 =  – xn+1 = g() – g(xn) = g(n) ( – xn), where n lies between xn and , that is, en+1 =
g(n)en.

Thus, we have en+1  k en. Using this repeatedly

en kn e0
lim en= 0 since k < 1,
m 

i.e. mlim

 – xn = 0  {xn}  

(The sequence {xn} converges to the number )

Hence proved.
Thus, it may be noted that the iterative scheme xn+1 = g(xn) converges under the condition g(x) < 1,
x  [a, b].

Example 2
For x3 – x – 1 = 0, find a positive root by the fixed point method. Find minimum number of iterations
so that nth approximate xn is correct to 4 decimal places.

Solution

1
Write x = (1  x) 3 = g(x).

The root lies between 1 and 2 since f(1) = –1 and f(2) = 3.

1
Also g(1) – 1 = – 1 = +ve
23
1
g(2) – 2 = 3 3 – 2 = –ve

I = [a, b] = [1, 2]
1
g(x) = 2 is decreasing function and
3(1  x) 3
 1
max g(x)= g(1) = 2 = k < 1.
xI
3 2 3
Since g(x) = +ve, therefore g(x) is increasing.
max g(x) = g(2) = 1 = 1.442 < 2
xI 33
1
min g(x) = g(1) = > 1.
xI 23
Hence, g(x)  I for all x  I.
1
Therefore, xn+1 = (1 x ) 3 generates a sequence of numbers which converges to a fixed point of g(x),
n
(starting with xo  I).
1
We have k = 2 < 1 and
3 2 3
en  kn en and eo< 1. Hence for the desired accuracy we have
n
 
 1  1
en   2
 < 10–4  n = 7.
  2
 3 2 3 

Remark 2: In the following figures we observe the importance of g(x) in the neighbourhood of a
fixed point .

y y
y = g(x) y=x y=#
y=x

y = g(x)
y = g(x)
x6 -x3 x1 x6 x2 x -x3 x1 x6 x2

Figure 3

In the neighbourhood of ,g(x)>1 (the sequences converge in these cases Fig. 3).
In the neighbourhood of , g(x) < 1 (the sequences converge in these two cases Fig. 4).
y y y=x
y=x

y = g(x)

y = g(x)

x
-x3 x1 x6 x2 x x2 x0

Figure 4

Remark 3: In numerical problems, one may follow the following procedure to find an interval [a, b].
In order to use this method one needs only to see if g(x)<1 at a point in the neighbourhood of the
root. Therefore, determining an interval I is not necessary.

Choose an interval [a, b] by some trial and check for the following:
(i) a – g(a) and b – g(b) must be of opposite sign (with b – g(b) > 0).
(ii) g(x)  k < 1 for x  [a, b]
(iii) g(x) is continuous on [a, b].

If above conditions are not satisfied try for a smaller interval and so on.

Example 3

Find the smallest positive root of e–x – cos x = 0 by the fixed point method.

Solution

To locate the smallest positive root, we draw the figures of

y = e–x and y = cos x

y
y=x y
y=x
(0, b)
1
.8
.6
(0, a) .4
.2 y=0

(0, 0) (a, 0) (b, 0) 1 x

Figure 5 Figure 6

π π
Figure shows that the desired root lies between 0 and i.e. in (0, ).
2 2
Now let us try x = cos–1 (e–x) = g(x)
1
g' (x) =
2x
e 1

To make this less than 1, we must choose e2x – 1 > 1, that is, e2x > 2. This gives
1
x> ln2 This suggest that we should take the suggested interval
2
1 π π
( ln2, ), but to take a closed interval, we consider I = [ln2, ].
2 2 2

Derivative of g(x) implies that g(x) is an increasing function.

g(x) = g(ln2) = cos–1 (e–ln2)
1 π 22
= cos–1   = = > ln2
2 3 21
 g(x) = g( π ) = cos–1 (  π ) < π
2 e 2 2
π
since e  2 is positive. Hence g(x)  I for all x  I.

 g' (x)  = 1 = 1 = 1 = k <1

2ln2 ln 4 3
e 1 e 1
since g' (x) is a decreasing function.

Hence all the sufficient conditions of the Theorem 4 are satisfied.

Now further, suppose that we want to find minimum number of iteration required to get 4 decimal
place accuracy. Let n be the minimum number of iterations required for the desired accuracy.
1
en  k n e o  10 4
2
π
eo  2  ln2  1. Thus the given condition is satisfied if
n
 1  1



  10  4 . That is,
 3 2
– n log 10 3  4  log 10 2 i.e.
4.301
n  18.07 i.e., n =19
0.238

Example 4: Find the iteration function and interval I = [a, b] which satisfy the conditions of the
theorem of fixed point to find the smallest positive root of
x = tan x.

Solution:

We rewrite the equation x = tan x as x  nπ  tan 1 x, n  0,1,2, ..

π π
We know that   tan 1 x  , so for desired root we take n = 1, that is,
2 2
We consider x = π  tan 1 x  g(x) and
 π 3π 
consider I =  , 
2 2 

 π 3π  π π π 3π
For x  1
 2 , 2  , we have   tan x  , hence  g(x) 
  2 2 2 2
1 1 4
  1
Also max.
xI
g' (x)  max. 1  x 2
xI π 2 4π2
1
4
(Since g' (x) is a decreasing function).
 π 3π 
Hence for any xo  I =  ,  , the sequence generated by the fixed-point iteration method will
2 2 
converge.

Remark 5: If ξ is a fixed point of g(x) lying in the open interval (c, d) on which φ' (x) then
the sequence {xn}generated with g(x) as iteration function will not converge to ξ , however close xo
to ξ is taken except accidentally. (Consider the root ξ = 2 of f(x)  x 2  x  2  0 with
g(x)  x 2  2 ).
Remark 6: If ξ is a fixed point of g(x) such that g' (ξξ  1, then the iteration function with g(x)
may or may not converge to  . However, if g' (ξξ  1, in some deleted neighbourhood of ξ ,
then it will converge to ξ , with xo taken sufficiently close to ξ . If g' (ξξ )  1, in some deleted
neighbourhood of ξ , then sequence will not converge to ξ .

Remark 7: The conditions mentioned in fixed-point theorem are sufficient but not necessary.

Now we discuss one example, which is very simple but conveys the fact that if a function f(x) has
more zeros i.e. f(x) = 0 has more than one real root, then we may have to consider different g(x) for
different roots.

Example 5: Find the iteration function g(x) and corresponding interval to get the two roots 1 and 2 by
fixed point iteration method for the equations
x2 – 3x + 2 = 0

Solution:

(a) For the root x = 1 if we consider x = 3x  2 = g(x), then

3
g(x) = and
2 3x  2
x2  2 1 5
g(1) = 1. Hence we choose g(x) = , I1 2 , 4
3  
2x 5
g(x) =
3
> 0 for x  I1. Hence g(x) is increasing. Also max.
xI 1
g' (x)  < 1.
6
25
max. g(x) = 16  2 57 < 5
xI 1  4
3 48
1
max. g(x) = 4  2 3 < 1
xI 1  2
3 2
Hence all the conditions of the theorem are satisfied.

(b) Now for the other root 2, consider

x2  2 4
If g(x) = , then g(2) =  1. Hence we choose g(x) = 3x  2 with
3 3
3 5
I2   , 
2 2
3
g(x) = > 0 for all x  I2, so g(x) is increasing.
2 3x  2
max. g(x) = 11 5
xI 2 
2 2
min g(x) = 5 3
xI 2  so g(x) : maps I 2 into itself.
2 2

3 3
g 1 (x)   1
Also max.
xI 9 10 (since g(x) is a decreasing function).
2 2 2
2

Hence all the conditions for the fixed point theorem are satisfied.
In the following two examples, we use the corollary to the fixed point theorem (Theorem 4).

Example 6: The equation f(x) = x4 – x – 10 = 0 has a root in the interval [1, 2]. Derive a suitable
iteration function (x) such that the sequence of iterates obtained from the method x k+1 =  (xk), k = 0,
1, 2, ... converges to the root of f(x)=0. Using this method and the initial approximation x 0 = 1.8,
iterate thrice.

1
Solution: Choose  (x) = (x  10) 4 , I = [1, 2].

1 3 1 1
Then (x) = 4
 x  10  
4  
3
4
 x  10  4
1 1
max. ‌(x)‌   3
x I2 4
(11) 4
i.e., ‌(x)‌  1 for x  [1, 2]

Given x0 = 1.8
x1 = 1.8  10  1 / 4 = 1.8534 = 1.86
x2 = 1.8  10  1 / 4 = 1.8534 = 1.86
x3 = 1.8  10  1 / 4 = 1.8534 = 1.86

Example 7: The equation f(x) = x3 – 5x +1 = 0 has a root in the interval [0, 1]. Derive a suitable
iteration function (x), such that the sequence of iterates obtained from the formula x k+1 = (xk), k = 0,
1, 2, ... converge to the root of f(x)=0. Using this formula and the initial approximation x 0 = 0.5,
iterate thrice.

x 3 1 3x 2 max
Solution:  (x) = is chosen since (x) = and 0 x I ‌(x) ‌< 1.
5 5
With x0 = 0.5, x1 = 0.225 = 0.23, x2 = 0.202.
1
What about choosing (x) = 5x  1  3 ?

What is 0max
x I
‌(x)‌ in this case?

2.3 CHORD METHODS FOR FINDING ROOTS

In the previous section we have introduced you to two iterative methods for finding the roots of an
equation f(x) = 0, namely bisection method and fixed point method. In this section we shall discuss
three iterative methods: regula-falsi, Newton-Raphson, and Secant methods. In the next section we
shall discuss the efficiency of these methods.

2.3.1 Regula-falsi Method

This method attempts to speed up bisection method retaining its guaranteed convergence. Suppose we
want to find a root of the equation f(x) = 0 where f(x) is a continuous function. We start this
procedure also by locating two points xo and x1 such that f(xo)f(x1) < 0.

Let us consider the line joining (xo, f(xo)) and (x1, f(x1)). This line cuts the x-axis at some point, say
x2. We find f(x2). If f(x2)f(x0) < 0, then we replace x1 by x2 and draw a straight line connecting (x2,
f(x2)) and {xo, f(xo)}. If f(x2) and f(xo) are such that f(x2)f(x0) > 0, then x0 is replaced by x2 and draw
a straight line connecting (x1, f(x1)) and (x2, f(x2)). Where the straight line crosses x – axis. that print
gives x3. In both the cases, the new interval obtained is smaller then the initial interval. We repeat
the above procedure. Ultimately the sequence is guaranteed to converge to the desired root.
y

x0 x2 x3 x4
x1

y = f(x)

Figure 7

f(x1 )  f(x o )
The equation of the chord PQ is y – f(x0) = (x  x o )
x1  x o
This cuts x-axis at the point x2 given by
f(x1 )  f(x o )
0 – f(xo) = (x 2  x o )
x1  x o
x o f(x 1 )  x 1 f(x o )
i.e. x2 = .
f(x 1 )  f(x o )
x r 1 f(x r )  x r f(x r 1 )
In general, xr+1 = , r = 1, 2, 3,....
f(x r )  f(x r 1 )

If f(x2) = 0, then x2 is the required root. If f(x2)  0 and f(xo)f(x2) < 0, then the next approximation
lies in (xo, x2). Otherwise it lies in (x2, x1). Repeat the process till x i 1  x i  ε .

Example 8: The equation 2x3 + 5x2 + 5x + 3 = 0 has a root in the interval [–2, –1]. Starting with xo =
–2.0 and x1 = –1.0 as initial approximations, perform three iteration of the Regula-falsi method.

Solution:
f(–2) = –16 + 20 – 10 + 3 = –3
f(–1) = –2 + 5 – 5 + 3 = 1, and f(–2) f(–1) < 0

x o f(x 1 )  x 1 f(x o ) 2  1  ( 1)( 3) 5

x2 = =  , i.e.,
f(x 1 )  f(x o ) 1  ( 3) 4
x2 = –1.25 (First iteration)
125 25 5 21
f(x2) =  5  5 3 
32 16 4 32

The root lies in (xo, x2)

x3 = –1.384 or 1.39 (Second iteration) since
 21  5  42  120
 2      ( 3)  42  5 
32  4 
x3 =  32 4  32
21 21 21  96
 ( 3) 3
32 32 32
162
=  1.384  1.39.
117

For next iteration find f(x3) and proceed in similar fashion.

2.3.2 Newton-Raphson Method

Newton-Raphson method or N–R method in short.

It can be introduced by basing it on the Taylor’s expansion as explained below. Let x o be an initial
approximation and assume that xo is close to the exact root  and
f’(xo)  0. Let  = xo + h where h is a small quantity in magnitude. Let f(x) satisfy all the conditions
of Taylor’s theorem. Then
f(xo+ h) = f(xo) + h f’(xo) + .....

The method is derived by assuming that the term involving h2 is negligible and that f(xo) + h f(xo) =
0 i.e. f(xo) + ( - xo)f(xo) = 0
f ( xo )
i.e.   xo -
f ' ( xo )
f ( xo )
i.e. x1 = xo -
f ' ( xo )
Geometrically the next approximation, x1, is the abscissa of the point of intersection of the tangent PT
and the x-axis in Figure 8.

The iteration scheme is

f(x n )
. xn+1 = xn – n = 0, 1, 2, ....
f' (x n )
y

x2 x1 x0

y = f(x)

Figure 8 : Newton – Raphson Method

N-R method is an extremely powerful technique, but it has a major difficulty – the need to know the
value of the derivative of f at each approximation or iteration. derivative evaluation, we discuss a
slight variation, known as Secant Method next.
Example 9: Newton-Raphson method is used to find the pth root of a positive real number R. Set
up the iteration scheme. Perform three iterations of the method for R=16.0, p =3, starting with the
initial approximation 2.0.

Solution: Let us denote pth root of R by x i.e.

x = R 1/p or xp ─ R = 0.
f (x) = pxp─1.

Newton-Raphson Iteration scheme is

f  xk 
xk+1 = x k  ,
f '  xk 
x kp  R
= xk  ,
px kp 1
 1 R
On simiplification we get xk+1 =  1  
 xk  , k  0, 1, 2, .....
 p p x kp 1
8 91
For R = 16, p = 3, xo = 2, we get x1 = = 2.67, x2 = = 2.53,
3 36

Remark 8: If a root is repeated m times, the N─R method is modified as

f  xk 
x k 1  x k  in
f  x k 

Example 10 : The quadric equation x4 – 4x2 + 4 = 0 has a double root. Starting with xo = 1.5,
compute two iterations by Newton-Raphson method.

Solution: For m-repeated root of f(x) = 0, the iteration scheme in case of Newton-Raphson method is
given by:

f  xk 
xk+1 = x k  m , k = 0, 1, 2, ...
f'  x k 
In this case, we have

xk+1 = x k 

2 x k4  4x k2  4  , (since m =2 and f(x) = x 4  4x 2  4 )
4x k3  8x k
With xo = 1.5, we have
1
2
3 16  17  1.41
x1 = 
2 3 12
2

Example 11: Perform two iterations of the Newton-Raphson method to find an approximate value of
1
starting with the initial approximation xo = 0.02
15
Solution: Suppose we want to find the reciprocal of the number N.
1
Let f(x) = –N
x
1
Then f’(x) = and the iteration scheme is
x2
1
N
xk
xk+1 = x k  = 2xk – Nx k2 , k = 0, 1, 2...
1
x k2

In this case we xk+1 = 2xk – 15 x 2k , k = 0, 1,2. This gives x1 = 0.034, x2 = 0.051,

x3 = 0.063, etc.

2.3.3 Secant Method

This method is a modification of the regula-falsi method and retains the use of secants throughout, but
dispenses with the bracketing of the root. Given a function f(x) and two given points x o and x1,

We compute,
f  xo 
xo   x o f  x1   x 1 f  x o 
x2 = f  x1   f  xo  .
f  x1   f  x o 
x1  x o
y

x3
x0 x2
x1

Figure 9

Above figure illustrates how xn+1 is obtained.

Example 12: Apply the Secant method to find a root of the equation

2x3 + 3x2 + 3x + 1 = 0. Take the initial approximations as xo = –0.2 and x1 = –0.4.

Solutions:

Let f(x) = 2x3 + 3x2 + 3x + 1

f(–0.2) = 0.504
f(–0.4) = 0.152
x o f  x1   x1 f  x o 
x2 = , = .1712 = 00.48
f  x1   f  x o  .352
You may now solve the following exercises:

E1) Using the Newton-Raphson method, find the square root of 10 with initial approximation x o =
3.

E2) A fixed point iteration to find a root of 3x3 + 2x2 + 3x + 2 = 0 close to

2  3x k  2x k2
xo = –0.5 is written as xk+1 = –
3x k2
Does this iteration converge? If so, iterate twice. If not, write a suitable form of the iteration,
show that it converges and iterate twice to find the root.

E3) Do three iterations of Secant method to find an approximate root of the equation.
3x3 – 4x2 + 3x – 4 = 0
Starting with initial approximations xo = 0 and x1 = 1.

E4) Do three iterations of fixed point iteration method to find the smallest positive roots of x 2 – 3x
+ 1 = 0, by choosing a suitable iteration function, that converges. Start with x o = 0.5.

E5) Obtain the smallest positive root of the equation of x3 – 5x + 1 = 0 by using 3 iterations of the
bisection method.

E6) Starting with xo = 0, perform two iterations to find an approximate root of the equation x 3 – 4x
+ 1 = 0, using Newton-Raphson method.

E7) Do three iterations of the Secant method to solve the equation

x3 + x – 6 = 0,
starting with xo = 1 and x2 = 2.

E8) Apply bisection method to find an approximation to the positive root of the equation.
2x – 3 Sinx – 5 = 0
rounded off to three decimal places.

E9) It is known that the equation x3 + 7x2 + 9 = 0 has a root between –8 and
–7. Use the regula-falsi method to obtain the root rounded off to 3 decimal places. Stop the
iteration when x i +1 x i < 10 4

E10) Determine an approximate root of the equation

Cosx – xex = 0
using Secant method with the two initial approximations as x o = 0 and
x1 = 1. Do two iterations

2.4 ITERATIVE METHODS & CONVERGENCE CRITERIA

Let {xn} be a sequence of iterates of a required root  of the equation f(x) = 0, generated by a given
method.

The error at the nth iteration, denoted by en is given by

e n =  – xn
The sequence of iterates {xn} converge to  if and only if en  0 as n   otherwise the sequence
of iterates diverges.
For each method of iteration considered by us, we shall discuss conditions under which the iteration
converges.

Let x0, x1, x2, etc be a sequence generated by some iterative method.

2.4.1 Order of Convergence of Iterative Methods

Definition 2: If an iterative method converges, that is, if {xn} converges to the desired root α , and
two constants p  1 and C > 0 exist such that

lim en+1
= C (C does not depend on n)
n →∞ enp

then p is called the order of convergence of the method and C is called the asymptotic error constant.
An iterative method with higher order of convergence than 1 is expected to converge rapidly. If p =
1, 2, 3, …, then the convergence is called linear, quadratic, cubic… respectively.

(i) For the Fixed Point Iteration method the order of convergence is generally 1, that is, it is of
first order (convergence is linear).
(ii) For the Newton-Raphson method, with xo near the root, the order of convergence is 2, that is,
of second order (convergence is quadratic).
(iii) For the Secant Method order of convergence is 1.618  1.62 but it is not guaranteed to
converge.
The bisection method is guaranteed to converge, but convergence is slow. Regula-falsi method is
guaranteed to converge. However, it is slow and order of convergence is 1.

2.4.2 Convergence of a Fixed Point Method

Theorem 5:

If g’(x) is continuous in some neighbourhood of the fixed point  of g, then the fixed point method
converges linearly provided g’(  )  0.

Proof: en+1 =  – x n+1

= g(  ) – g(xn)
= g’(  n) (  – xn)
for some  n lying between xn and  .
en+1 = en g’(  n)
Since g’(x) is continuous in a neighbourhood of  , we can write
g’(  n) = g’(  ) + hn such that lim hn = 0.
n →∞
en+1 = en{g’(  ) +hn}

On taking n   , we have

e n 1
lim  g'  ξ  = C  0
n  en

e n 1
Hence p = 1, since  g'  ξ 
en
Therefore, fixed point method converges linearly.

Note: Smaller the value of g'  ξ  , faster would be the convergence.

Theorem 6: If g”(x) is continuous in some neighbourhood of the fixed point  of g, then the fixed
point method converges quadratically, provided g’(  ) =0 and g”(  )0.

Proof: We have

en+1 =  – xn+1
en+1 = g(  ) – g(xn)
= g(  ) – g(  – en)
By using Taylor’s series expansion, we have
en+1 = g(  ) – {g(  ) – en g’(  ) +
2
en
g"  η n  
2

for some  n lying in the interval of xn and  . That is,

en+1
e2
 
= - n g ''  ξ   hn since g”(x) is continuous, hn  0. as
2
n  .

e n 1 g"  ξ 
Thus, nlim 2
 = C0
 en 2

Here p = 2, hence convergence is quadratic.

Fixed-point iteration is effective when it converges quadratically, as in Newton-Raphson method

discussed below.

N─R Method

We define for equation f(x) = 0 an iterative function g(x) as

f  x
g(x) = x – , then the method is called Newton’s method. We state a theorem without
f'  x 
proof which suggests an interval in which if xo is taken then Newton’s method converges. We
f  xn 
generate the sequence {xn}as xn+1 = xn – , n =1, 2, 3, …
f'  x n 

2.4.3 Convergence of Newton’s Method

Theorem 7:

Suppose we are to solve f(x) = 0. If f’(x)  0 and f”(x) is continuous on the closed finite interval [a,
b] and let the following conditions be satisfied:

(i) f(a) f(b) < 0

(ii) f”(x) is either  0 or  0 for all x  [a, b]
(iii) At the end points a, b

f  a f  b
< b – a and < b – a.
f'  a  f'  b 
Then Newton’s method converges to the unique solution  of f(x) = 0 in [a, b] for any choice of x0 
[a, b].

Theorem 8: Let f(x) be twice continuously differentiable in an open interval containing a simple root
 of f(x) = 0. Further let f"(x) exists in neighbourhood of. Then the Newton’s method converges
quadratically.

f  x
Proof: g(x) = x – is continuously differentiable in some open neighbourhood of  . On
f '  x
differentiating g(x), we get
 f'  x   2  f(x)f'  x 
g (x) = 1 –
 f'  x   2
=
f( x)f"  x 
 f'  x   2

g (  ) = 0, since f(  ) = 0 (f’(  )  0 for a simple root  )

 2f(x) f'  x   f"  x     f'  x    f'  x  f"  x   f  x  f" '  x 
2 2

Also g”(x) =  f'  x   4

 f'  x 2
f"  x  f  x  f'  x  f" '  x   2f  x   f"  x 2

=  f'  x 3
g”(  ) =
f"  ξ   0
f'  ξ 

By Taylor’s formula, we have

en+1 =  – xn+1 = g(  ) – g(xn)
1
= –g’(  )(xn–  ) – g”(  n)(xn–  )2 for some  n between  and xn. That is
2
1
en+1 = g’(  )en – g”(  n)e2 n
2
since g’(  ) = 0, and g”(x) is continuous, we have
1
en+1  – g”(  ) e n .
2

Hence the Newton’s Method converges quadratically if of x o is chosen sufficiently close to  , where
 is a simple root of  .

2.4.4 Rate of Convergence of Secant Method

Suppose f(x) = 0 is to be solved. Consider the curve y = f(x).
y

C
x0 x3 x2

x3
(x2 f(x9))
 Figure 10

Let the chord AB through the points A(xn–1, f(xn–1)) and B (xn, f(xn)) be drawn. Suppose this intersects
x–axis at C. Denote this value of x by xn+1. That is
f(x n )  f(x n 1 )
y – f(xn-1) = (x – xn–1)
x n  x n 1
y = 0, x = xn+1, we get

f(x n 1 )(x n  x n 1 )
xn+1 = xn–1 –
f  x n   f  x n 1 

x n 1 f(x n )  x n f(x n 1 )
=
f  x n   f  x n 1 

This is known as secant method. The sequence {xn} is generated with starting points x0, x1. We get
x2, reject x0, and use x1, x2 to get x3 and so on.

Let en+1 =  – xn+1

x n 1 f(x n )  x n f(x n 1 )
= –
f  x n   f  x n 1 
writing en =  – xn, en–1 =  – xn–1
and xn =  – en, xn–1 =  – en–1

we get
e n 1 f(ξ  e n )  e n f(ξ  e n 1 )
en+1 =
f  ξ  e n   f  ξ  e n 1 

 en2 
= e f(ξ(e f'(ξξ 2 f"(ξξ higher order terms
n1 n

f(  ) = 0
 e2 

since
 en  f(ξ(  en 1 f' (ξξ  n f" (ξξ  higher order terms 
 2 
f(ξ(  e n f' (ξξ f(ξ(  e n1 f' (ξξ

en+1
1
2
e 2
n e n 1  e n e 2
n 1  f"

  e n  e n 1  f' (ξξ

1 f" (ξξ
=  enen–1, for sufficiently large n
2 f' (ξξ

1 f" (ξξ
Let 
2 f' (ξξ
=  (fixed constant).
Then en+1 =  enen–1

Suppose en+1 = C e np  e n  Ce np1 substituting these in previous results

1
e p 1
=  en  n   1 e n 1 p
α
Ce np
C C p
Equating powers of p, on both sides we get
12
P = 1 + p  p  p 1  0
 1 5
P= , p = 1.618.
2

Now after comparing the rate of convergence of fixed point method, N─R Method and Secant method,
we find that secant is faster than fixed point month and N─R method is faster than secant method. For
further qualitative comparison refer the books mentioned.
Apart from the rate of convergence, the amount of computational effort required for iteration and the
sensitivity of the method to the starting value and the intermediate values, are two main basis for
comparison of various iterative methods discussed here. In the case of Newton’s method, if f  (x) is
near zero anytime during the iterative cycle, it may diverge. Furthermore, the amount of
computational effort to compute f(x) and f  (x) is considerable and time consuming. Whereas the
fixed point method is easy to programme.

You may now solve the following exercises.

E11) Let M denote the length of the initial interval a0 ,b0  . Let  x 0 , x1 , x3 ... represent the
M
successive midpoints generated by the bisection method. Show that x i 1  x i 
2 i2
Also show that the number n of iterations required to generate an approximation to a root to an
accuracy  is given by

n  2 
log ε  M
log2

E12) If x  ξ is a zero of f(x) of order 2, then f      , f '      and f "     

. Show that in this case Newton-Raphson’s method no longer converges quadratically. Also
show that if f '      , f "      and f " '  x  is continuous
in the neighbourhood of  , the iteration
2f(xi )
x i 1  x i   g(x i )
f(x i )
does converge quadratically.

E13) The quadratic equation x 4  4x 2  4  0 has a double root at 2 . Starting with x 0  1.5,
compute three successive approximations to the root by Newton-Raphson method. Do this
f(x) 2f(x)
with g 1 (x)  x  and g 2 (x)  x  and comment on the order of convergence
f' (x) f' (x)
from your results.

E14) The following are the five successive iterations obtained by the Secant method to find the real
positive root of the equation x 3  x  1  0 starting with x 0  1.0 and x 1  2.0.

n 2 3 4 5 6 7
x n 1.166667 1.2531120 1.3372064 1.3238501 1.3247079 1.3247180

Calculate e n and e n  e n e n  for n = 2, 3, 4. Also compute the constant directly

 f' '  ξ  
 2f'  ξ   assuming the value of correct to eight decimal places


as  = 1.324718.
E15) If a 0  0 and b 0  1.0 , how many steps of the bisection method are needed to determine the
root with an error of at most 10 5 ?

2.5 SUMMARY
In this unit we have covered the following points:
The methods for finding an approximate solution of equation in one variable involve two steps:
(i) Find an initial approximation to a root.
(ii) Improve the initial approximation to get more accurate value of the root.

The following iterative methods have been discussed:

(i) Bisection method

(ii) Fixed point iteration method
(iii) Regula-falsi method
(iv) Newton-Raphson method
(v) Secant method

We have introduced the convergence criterion of an iteration process.

We have obtained the order/rate of convergence for the iterative methods discussed.
Finally we have given a comparative performance of these methods.

2.6 SOLUTIONS/ANSWERS
E1) x= 10 , i.e. x2 = 10. f(x) = x2 – 10
x n2  10 x n2  10
xn+1 = xn  = , n = 0, 1, 2,
2x n 2x n
19 (3.167) 2  10
xo = 3, x1 = = 3.167, x2 = = 3.162
6 6.334

E2) Here  ( x ) = –
1
2
2  3x  2x  2
3x
1  4  3x 
 ' ( x) = 1 at xo = –0.5
3 x3

Hence iteration does not converge.

If  ( x ) = –
1
3

2  2x 2  3x 3 , then 
 ' ( x) = 
1
3

4x  9x 2  1 at xo = –0.5

Hence in this case iteration converges

First iteration x1 = –0.708
Second iteration x2 = –0.646

E3) f(x) = 3x3 – 4x2 + 3x – 4, xo = 0, x1 = 1.

 x n 1 f  x n   x n f  x n 1 
xn+1 = , n = 1, 2, 3
f  x n   f  x n 1 
This gives x2 = 2, x3 = 1.167, x4 = 1.255

E4) Root lies in [0, 1]. We take.

x 2 1
x  g(x)
3
2x
g'  x    g'(x)  1
3

Starting with x0 = 0.5, we have

5 169
x1   0.417, x 2   0.391 and x 3  0.384
12 432

E5) f(0) > 0 and f(1) < 0. The smallest positive root lies in ]0, 1[.

No. of bisection Bisected value f(xi) Improved

xi interval
1 0.5 – 1.375 ]0,0.5[
2 0.25 – 0.09375 ]0,0.25[
3 0.125 0.37895 ]0.125, 0.25[

It is enough to check the sign of f(x0) – the value need not be calculated.

The approximate value of the desired root is 0.1875.

E6) Here f(x) = x3 – 4x + 1, x0 = 0.

f (xx )
The iteration formula is xi 1  x 2 
f 1 ( xi )
2xi3  1
i.e. xi 1  .
3xi2  4

This gives
x1  0.25, x1  0.254095  0.2541

E7) f ( x)  x 3  x  6, x 0  1, x1  2
x n 1 f ( x n )  x n f ( x n 1 )
x n 1  n = 1, 2, 3, …
f ( x n )  f ( x n 1 )

This gives x 2  1.5, x3  1.609  1.61, x 4  1.64.

E8) Here f(x) = 2x – 3 Sinx – 5

x 0 1 2 2.5 2.8 2.9

f(x) – 5.0 – 5.51224 – 3.7278 – 1.7954 – 0.4049 0.0822

Thus a positive root lies in the interval [2.8, 2.9].

 No. of bisection Bisected value x 0 f  x0  Improved Interval
1 2.85 –0.1624 [2.85, 2.9]
2 2.875 –0.0403 [2.875, 2.9]
3 2.8875 –0.02089 [2.875, 2.8875]
4 2.88125

x0 f(x1 )  x1 f(x0 ) ( 8).9  ( 7)( 55)

E9) x2    7.1406
f(x1 )  f(x0 ) 0  55

Similarly x 3  7.168174
The iterated values are presented in tabular form below:

No. of Interval Bisected The function

intersections value x 0 value f  xi 
1 ] –8, –7[ –7.1406 1.862856
2 ] –8, –7.1406[ –7.168174 0.358767
3
4
5
6

Complete the above table. You can find that the difference between the 5 th and 6th iterated
values is 7.1748226  7.1747855  0.0000371 signaling a stop to the iteration. We
conclude that –7.175 is an approximate root rounded to the decimal places.

E10) Here f(x)  cosx  xe 2 , x 0  0 and x1  1

x0 f(x 1 )  x 1 f(x 0 )
x2   0.3146653378
f(x 1 )  f(x 0 )
x 1 f(x 2 )  x 2 f(x1 )
x3   0.4467281466
f(x 2 )  f(x1 )

E11) Starting with bisection method with initial interval  a 0 , b0  (recall that in each step the
1
interval width is reduced by we have
2
b  a0 M
b1  a 1  0 
2 2

b1  a1 b0  a0
b2  a 2  
2 2
b0  a0
and finally bn  a n 
2n
a  bn
Let x n  n .
2

a n  bn
2

xn
 an bn

xn+1

a n 1  b n 1 a n  b n
Then x n 1  x n  
2 2
a n  bn
We have either a n 1  and bn 1  bn
2
a  bn
or a n 1  a n and bn 1  n .
2
In either case

bn  a n b0  a 0
x n 1  x n  
22 2 n2

M
We want x n 1  x n  ε.
2 n 2

This is satisfied if

 M 
log  n  2   logε
2 

log M – (n + 2) log 2 < log 

nlog2 > – 2log2 + log M – log 

log ε  M
n  2 
log2

E12) In case we have for a simple root 

f(x n )
x n 1  x n   g xn 
f' (x n )
f( x)f" (x)
g' ( x) 
 f' ( x) 2

. Thus
f" (ξξ
g' (ξξ  0 since f(  ) = 0 and g" (ξξ  (f' (ξξ  0)
f' (ξξ
But given that f(  ) = 0 = f’(  ) and f”( )  0.

In this case
 lim
x ξ
f( x)f " (x )
 f' ( x) 2 =
f(x)
lim lim f " ( x)
x 
 f' (x) 2 x 

f' (x) 1
= lim
x 
lim
x 
f" (x) = (by L’Hospital Rule)
2f(x)f" (x) 2
1
That is g' (ξξ  0
2
Hence it does not converge quadratically.

In case

2f(x n ) 2f(x)
x n 1  x n  where g(x) = x 
f' (x n ) f' (x)
g ' ( x)

2f( x) f" (x )   f' (
 f' ( x) 2

and g' (ξξ  0 .

Since
g’(x)
2f(x)f" (x )
  1
 f' (x)
2

and
lim g’(x) = lim
x  x 
2f( x)f" (x)
 1
 f' ( x) 2

2f(x)
= lim  lim
x  f”(x) – 1
x 
 f' (x) 2
1
 2  1  0.
2

E13) f(1.5) = 5.0625 – 9 + 4 = .0625

f’(1.5) = 13.5 – 12 = 1.5

With g1 ( x)
.0625
x1 = 1.5   1.5 – .04 = 1.46
1.5
f(1.46) = 4.543 – 8.52 + 4 = 0.02
f’(1.46) = 12.45 – 11.68 = 0.77
0.02
x2 = 1.46   1.44
0.77

With g 2 ( x )
2 .0625
x1 = 1.5   1.5 – 0.08 = 1.42
1.5
f(1.42) = 4.065 – 8.065 + 4 = 0
x2 = 1.42
 Actual root = 1.4142. Hence convergence is faster with g 2 ( x ) with two decimal digit
arithmetic.

E14) We have the following results in tabular form:

n 1 2 3 4 5
en 0.1580513 0.0716060 0.012884 0.0008679 0.0000101
e n 1 e n e n 1 1.1034669 0.9705400 0.9318475

Also f" (ξξ 2f' (ξ ) = 0.93188  ξ  1.3247180 

e5 e 4 e 3  0.9318475. Hence agreement is good for n = 4.

E15) Here M = b0  a0  1  0  1
log ε  M
n  2 
log2
Here  = 10 5
log10 5
So n  2 
log2

5log10
 2 
log2
5
 2 
log2
 2  5  3.322
 2  16.61
 14.61
n  15