See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332632665

New Three-Steps Iterative Method for Solving Nonlinear Equations

Article · April 2019

CITATIONS READS

4 288

2 authors, including:

Oghovese Ogbereyivwe
Delta State University
19 PUBLICATIONS 24 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

High order quadrature based View project

Efficient High order quadrature based iterative methods for approximating nonlinear system of equations. View project

All content following this page was uploaded by Oghovese Ogbereyivwe on 25 April 2019.

The user has requested enhancement of the downloaded file.

IOSR Journal of Mathematics (IOSR-JM)
e-ISSN: 2278-5728, p-ISSN: 2319-765X. Volume 10, Issue 5 Ver. IV (Sep-Oct. 2014), PP 44-47
www.iosrjournals.org

New Three-Steps Iterative Method for Solving Nonlinear

Equations
1
Ogbereyivwe Oghovese, 2Emunefe O. John
1
(Department of Mathematics and Statistics, Delta State Polytechnic, Ozoro, Nigeria)
2
(Department of General Studies, Petroleum Training Institute, Effurun, Delta State, Nigeria)

Abstract: In this paper, we proposed a new three steps iterative method of order six for solving nonlinear
equations. The method uses predictor–corrector technique, is constructed based on a Newton iterative method
and the weight combination of mid-point with Simpson quadrature formulas. Several numerical examples are
given to illustrate the efficiency and performance of the iterative methods; the methods are also compared with
well known existing iterative method.
Keywords: Newton method, Order of convergence, Predictor-Corrector method, Quadrature method,.

I. Introduction
The nonlinear equation, often arise from the numerical modeling of problems in many branches of science and
engineering [1]. These equations more often are not solved analytically hence resort to numerical solutions.
More robust and efficient methods for solving the nonlinear equations are continuously being sought. There are
many papers that deal with nonlinear algebraic equations, such as, improving Newton Raphson method for
nonlinear equations by modified Adomian decomposition method [2], iterative method improving Newton's
method by the decomposition method [3], a third-order Newton-type method to solve system of nonlinear
equations [4], a variant of Newton's method with accelerated third-order convergence [5], variants of Newton’s
Method using fifth-order quadrature formulas [6] and other methods (see [7-16] and reference therein).
In this paper, we recommend an iterative method to solve nonlinear equations. We show that our
proposed method is of convergence order six. The obtained results suggested that the proposed methods
introduce a powerful improvement for solving nonlinear equations.

II. Development of the methods

Consider a nonlinear equation
𝑓 𝑥 =0 (1)
If we write the differentiable function 𝑓(𝑥) as follows
𝑥

𝑓 𝑥 = 𝑓 𝑥𝑛 + 𝑓 ′ (𝑡)𝑑𝑡 (2)
𝑥𝑛
and then approximate the integration in (2) with average of midpoint and Simpson quadrature formulas then we
have
𝑥
𝑥 − 𝑥𝑛 ′ 𝑥 − 𝑥𝑛 𝑥 − 𝑥𝑛 ′ 𝑥 − 𝑥𝑛
𝑓 ′ (𝑡)𝑑𝑡 = 𝑓 + 𝑓 𝑥 + 10 + 𝑓 ′ 𝑥𝑛 (3)
2 2 12 2
𝑥𝑛
From (2) and (3), we have
𝑥 − 𝑥𝑛 𝑥 − 𝑥𝑛
𝑓 𝑥 = 𝑓 𝑥𝑛 + 𝑓 ′ 𝑥 + 10𝑓 ′ + 𝑓 ′ 𝑥𝑛 (4)
12 2
Since 𝑓 𝑥 = 0 then
12𝑓(𝑥𝑛 )
𝑥 = 𝑥𝑛 − 𝑥 − 𝑥𝑛 (5)
𝑓 ′ 𝑥 + 10𝑓 ′ + 𝑓 ′ 𝑥𝑛
2
Equation (5) is a fixed point formulation which enable us to suggest and analyze a three-step iterative method
for solving nonlinear equations (1).

Algorithm 2.1
INPUT initial approximation𝑥0 ; tolerance 𝜀, maximum number of iterations 𝑁0 .
OUTPUT approximate solution 𝑥𝑛+1 , or message of failure.

Step 1 : Set 𝑛 = 0 and 𝑖 = 1.

Staep2 : While 𝑖 ≤ 𝑁0 do steps 3-5.
www.iosrjournals.org 44 | Page
 New Three-Steps Iterative Methods for Solving Nonlinear Equations

Step 3 : Calculate
𝑓(𝑥𝑛 )
𝑦𝑛 = 𝑥𝑛 − (6)
𝑓 ′ (𝑥𝑛 )
12𝑓(𝑥𝑛 ) 𝑥 − 𝑦𝑛
𝑧𝑛 = 𝑥𝑛 − ′ , 𝑤𝑛 = (7)
𝑓 𝑥𝑛 + 10𝑓 𝑤𝑛 + 𝑓 𝑦𝑛
′ ′ 2
𝑓(𝑧𝑛 )
𝑥𝑛+1 = 𝑧𝑛 − ′ (8)
𝑓 (𝑧𝑛 )
Step 4 : If 𝑥𝑛 − 𝑥𝑛+1 < 𝜀, then OUTPUT 𝑥𝑛+1 ; stop
Step 5 : Set 𝑛 = 𝑛 + 1; 𝑖 = 𝑖 + 1 and go to step 2.
Step 6 : OUTPUT (‘Method failed after 𝑁0 iterations, 𝑁0 =’ 𝑁0 ); stop.

Algorithm 2.1 is called the Predictor-corrector Newton’s method (PCNM) and has sixth order convergence. At
each iteration point, Algorithm 2.1 requires two function evaluation and four function derivative evaluation. We
take into account the definition of efficiency index [12], if we suppose that all the evaluations have the same
1
cost as function one, we have that the efficiency index of Algorithm 2.1 is 66 ≈ 1.3480 which is better of than
1
the efficiency index of DHM method [16] and Mir et al. method [14] with 34 ≈ 1.316, Cordero and Torregrosa
1 1
method [6] with 35 ≈ 1.2457 and Rostam method [13] with 67 ≈ 1.2917.

III. Convergence Analysis

In this section, we establish the six-order convergence of the proposed method in this paper.

Theorem 3.1: Let 𝛼 be a simple zero of sufficiently differentiable function 𝑓: 𝑅 → 𝑅 for an open interval 𝐼. If
𝑥0 is sufficiently close to , then the three step method defined by Algorithm 2.1 has convergence at least of
order six.

Proof: Let 𝛼 be a simple zero of 𝑓, and 𝑒𝑛 = 𝑥𝑛 − 𝛼. Using Taylor expansion around 𝑥 = 𝛼 and taking into
account 𝑓 𝛼 = 0, we get

𝑓 𝑥𝑛 = 𝑓 ′ 𝛼 𝑒𝑛 + 𝑐2 𝑒𝑛2 + 𝑐3 𝑒𝑛3 + 𝑐4 𝑒𝑛4 + ⋯ (9)

𝑓 ′ 𝑥𝑛 = 𝑓 ′ 𝛼 1 + 2𝑐2 𝑒𝑛 + 3𝑐3 𝑒𝑛2 + 4𝑐4 𝑒𝑛3 + 5𝑐5 𝑒𝑛4 + ⋯ (10)

where
1 𝑓𝑘 𝛼
𝑐𝑘 = 𝑘 ! 𝑓 ′ , 𝑘 = 1,2,3, … and 𝑒𝑛 = 𝑥𝑛 − 𝛼
𝛼

From (9) and (10), we get

𝑓 𝑥𝑛
= 𝑒𝑛 − 𝑐2 𝑒𝑛2 + 2 𝑐22 − 𝑐3 𝑒𝑛2 + 7𝑐2 𝑐3 − 4𝑐23 − 3𝑐4 𝑒𝑛4 + 𝑂 𝑒𝑛5 … (11)
𝑓 ′ 𝑥𝑛

From (6) and (11) we have

𝑦𝑛 = 𝛼 + 𝑐2 𝑒𝑛2 + 2 𝑐3 − 𝑐22 𝑒𝑛2 + −7𝑐2 𝑐3 + 4𝑐23 + 3𝑐4 𝑒𝑛4 + 𝑂 𝑒𝑛5 … (12)

From (12), we get,

2 3 4
𝑓 𝑦𝑛 = 𝑓 ′ 𝛼 𝑦𝑛 − 𝛼 + 𝑐2 𝑦𝑛 − 𝛼 + 𝑐3 𝑦𝑛 − 𝛼 + 𝑐4 𝑦𝑛 − 𝛼 +⋯ (13)

and

𝑓 ′ 𝑦𝑛 = 𝑓 ′ 𝛼 1 + 2𝑐22 𝑒𝑛2 + 𝑐2 𝑐3 − 𝑐23 𝑒𝑛3 + −11𝑐22 𝑐3 + 8𝑐24 + 6𝑐2 𝑐4 + ⋯ 14

Expanding 𝑓 ′ 𝑤𝑛 about 𝛼, we get

3 7 1
1 + 𝑐2 𝑒𝑛 + 𝑐22 + 𝑐3 𝑒𝑛2 + 𝑐2 𝑐3 − 2𝑐23 + 𝑐4 𝑒𝑛3
𝑓 ′ 𝑤𝑛 = 𝑓 ′ 𝛼 4 2 2
9 29 2 5
+ 𝑐2 𝑐4 − 𝑐2 𝑐3 + 𝑐5 + 4𝑐24 + 3𝑐32 𝑒𝑛4 + ⋯ (15)
2 4 16
www.iosrjournals.org 45 | Page
 New Three-Steps Iterative Methods for Solving Nonlinear Equations

then
3
𝑓 ′ 𝑥𝑛 + 10𝑓 ′ 𝑤𝑛 + 𝑓 ′ 𝑦𝑛 = 12𝑓 ′ 𝛼 1 + 𝑐2 𝑒𝑛 + 𝑐22 + 𝑐3 𝑒𝑛2
4
13 3
3 3
17 53 65 5
+ 𝑐2 𝑐3 − 2𝑐2 + 𝑐4 𝑒𝑛 + 𝑐2 𝑐4 − 𝑐22 𝑐3 + 𝑐5 + 4𝑐24 + 𝑐32 𝑒𝑛4 + ⋯ (16)
4 4 4 8 96 2

From (7), (9) and (16), we get

1 27 1
𝑤𝑛 = 𝛼 + 𝑐3 𝑐22 − 𝑐32 𝑒𝑛3 + − 𝑐2 𝑐3 + 2𝑐22 + 𝑐4 𝑒𝑛4 + 𝑂 𝑒 5 (17)
8 8 4
2 3 4
𝑓 𝑧𝑛 = 𝑓 ′ 𝛼 𝑧𝑛 − 𝛼 + 𝑐2 𝑧𝑛 − 𝛼 + 𝑐3 𝑧𝑛 − 𝛼 + 𝑐4 𝑧𝑛 − 𝛼 +⋯

1 1 27 1 1
= 𝑓′ 𝛼 𝑐3 𝑐22 − 𝑐32 𝑒𝑛3 + 𝑐4 − 𝑐2 𝑐3 + 3𝑐22 𝑒𝑛4 + 𝑐33 𝑐23 − 𝑐2 𝑐34 − 𝑐32 𝑐25 𝑒𝑛6
8 4 8 4 64
1 3 5
27 2 2 1 2
27 2 4 3 2 3 7
+ − 𝑐3 𝑐2 𝑐4 − 6𝑐3 𝑐2 − 𝑐3 𝑐2 + 𝑐2 𝑐3 𝑐4 + 𝑐3 𝑐2 + 𝑐3 𝑐2 𝑒2 + ⋯ (18)
2 32 16 4 4
2 3
𝑓 ′ 𝑧𝑛 = 𝑓 ′ 𝛼 1 + 2𝑐2 𝑧𝑛 − 𝛼 + 3𝑐3 𝑧𝑛 − 𝛼 + 4𝑐4 𝑧𝑛 − 𝛼 +⋯

1 27 1 3 3
= 𝑓 ′ 𝛼 1 + − 𝑐2 𝑐32 + 2𝑐3 𝑐23 𝑒𝑛3 + − 𝑐22 𝑐3 + 𝑐2 𝑐4 + 6𝑐23 𝑒𝑛4 + 3𝑐33 𝑐24 + 𝑐35 − 𝑐34 𝑐22 𝑒𝑛6
4 4 2 64 4
81 3 3 3 3 81 9
+ − 𝑐2 𝑐3 + 18𝑐32 𝑐24 + 𝑐32 𝑐22 𝑐4 − 𝑐33 𝑐4 + 𝑐34 𝑐2 − 𝑐33 𝑐22 𝑒𝑛7 + ⋯ (19)
4 2 16 31 4

From (8), (18) and (19), we get

3 3 3 3
𝑥𝑛+1 = 𝛼 + 𝑐 𝑐 − 𝑐 4 𝑐 − 3𝑐32 𝑐25 𝑐𝑛6 + 𝑂 𝑒𝑛7 (20)
4 3 2 64 3 2
or
3 3
𝑒𝑛+1 = 4 𝑐33 𝑐23 − 64 𝑐34 𝑐2 − 3𝑐32 𝑐25 𝑐𝑛6 + 𝑂 𝑒𝑛7 (21)
which shows that Algorithm 2.1 has sixth-order of convergence.

IV. Numerical Example

In this section, we have worked out some examples to illustrate the efficiency of the proposed method, by
comparing the Newton method (NM), the method of Cordero and Terregrossa [6] (CTM), the method of Parhi
and Gupta [15] (PGM), and our method (PCNM). The following stopping criteria were used:
𝑖. 𝑥𝑛+1 − 𝑥𝑛 < 𝜀 𝑖𝑖. 𝑓 𝑥𝑛+1 < 𝜀
where 𝜀 = 10−15 .

The following examples, some which are from [6,13,15], were used.

𝑓1 𝑥 = 𝑥 3 + 4𝑥 2 − 10
𝑓2 𝑥 = 𝑠𝑖𝑛2 𝑥 − 𝑥 2 + 1
𝑓3 𝑥 = 𝑥 2 − 𝑒 𝑥 − 3𝑥 + 2
𝑓4 𝑥 = cos 𝑥 − 𝑥
𝑓5 𝑥 = (𝑥 − 1)3 − 1
𝑓6 𝑥 = 𝑥 3 − 10
2
𝑓7 𝑥 = 𝑥𝑒 𝑥 − 𝑠𝑖𝑛2 𝑥 + 3𝑐𝑜𝑠𝑥 + 5
1
𝑓8 𝑥 = 𝑥 − − 3 (22)
𝑥
𝑓9 𝑥 = 𝑒 𝑥 + 𝑥 − 20
𝑓10 𝑥 = 𝑥 3 − 𝑥 2 − 1
𝑓11 𝑥 = 𝑥 3 − 9𝑥 2 + 28𝑥 − 30
𝑓12 𝑥 = 𝑠𝑖𝑛𝑥 + 𝑥𝑐𝑜𝑠𝑥
𝑓13 𝑥 = 𝑒 −𝑥 + 𝑐𝑜𝑠𝑥
𝑓14 𝑥 = 𝑥10 − 1
𝑓15 𝑥 = (𝑥 − 2)23 − 1

www.iosrjournals.org 46 | Page
 New Three-Steps Iterative Methods for Solving Nonlinear Equations

Table 4.1: Comparison between other methods

𝑓(𝑥) 𝑥0 NM CTM PGM PCNM
𝑓1 1.6 5 4 3 3
𝑓2 1.0 6 4 3 3
𝑓3 1.5 6 4 3 3
𝑓4 4.0 30 6 12 3
𝑓5 1.5 8 6 4 4
𝑓6 4.0 7 5 4 4
𝑓7 -1.2 4 3 3 3
𝑓8 1.3 7 5 6 4
𝑓9 2.5 6 4 3 3
𝑓10 0.5 13 Div 9 9
𝑓11 0.1 8 6 5 4
𝑓12 3.0 6 4 4 3
𝑓13 3 8 5 4 3
𝑓14 3.5 18 11 6 6
𝑓15 5.0 31 20 9 9

V. Conclusions
We presented a new modification of Newton-type method with sixth-order convergence for solving nonlinear
equations. Application on some numerical examples shows that the new algorithm performs better in terms of
number of iterations to achieve solution to nonlinear equations than the classical Newton’s method and some
existing methods. Its efficiency is also better than some other methods.

Competing Interests
The authors declare that no competing interests exist.

Reference
[1] J.M. Ortega, W.C. Rheinboldt, Iterative Solution of Nonlinear Equations in Several Variables, Academic Press, New York, 1970.
[2] S. Abbasbandy, Improving Newton Raphson method for nonlinear equations by modified Ado-mian decomposition method, Appl.
Math. Comput, 145 (2003), 887{893.
[3] C. Chun, Iterative method improving Newton's method by the decomposition method, Comput. Math. Appl, 50 (2005), 1559{1568.
[4] M.T. Darvishi and A. Barati, A third-oredr Newton-type method to solve system of nonlinear equations, Appl. Math. Comput, 187
(2007), 630{635. Math. Comput, 169 (2004), 161{169.
[5] S. Weerakoon and T.G.I. Fernando, A variant of Newton's method with accelerated third-order convergence, Appl. Math. Lett
13(8)(2000) 87:93.
[6] Cordero, A. and Torregrosa, J. R., (2007). Variants of Newton’s Method using fifth-order quadrature formulas. Appl. Math.
Comput., 190:686-698.
[7] R.L. Burden and J.D. Faires, Numerical Analysis,7th edition, PWS Publishing Company, Boston,2001.
[8] C. Chun, Some variants of Kings fourth-order family of methods for nonlinear equations, Appl.Math. Comput, 190 (2007), 57{62.
[9] C. Chun, A family of composite fourth-order iterative methods for solving non-linear equations, Appl. Math. Comput, 187 (2007),
951{956.
[10] C. Chun, Some fourth-order iterative mthods for solving non-linear equations, Appl. Math. Comput, 195 (2008), 454{459.
[11] C. Chun, Y. Ham, Some sixth-order variants of Ostrowski root-_nding methods, Appl. Math. Comput. 193 (2007), 389{394.
[12] Ostrowski, A. M., (1966). Solutions of Equations and System of Equations, Academic Press, New York-London, (1966).
[13] Rostam, K. S., (2010). Six Order Iterative Method for solving Nonlinear Equations. World Applied Sciences Journal, 11(11):1393-
1397.
[14] Mir, N. A., Ayub, K. and Rafiq, A. (2010). “ A Third-Order Convergent Iterative Method for Solving Non-linear Equations”.
International Journal of Computer Mathematics, Vol. 87, No. 4, pp. 633-637.
[15] Parhi, S.K. and Gupta D.K., (2008). A sixth order method for nonlinear equations. Appl. Math. Comput., 203:50-55.
[16] Dehghan M. and Hajarian M., (2010). “Some derivative free quadratic and cubic convergence iterative formula for solving
nonlinear equations” Comp & App Math 29:19-30.

www.iosrjournals.org 47 | Page

View publication stats