Abstract
We study the convergence of an iterative method for solving the equation (fleft( xright) =0, f:Isubseteq mathbb{Rrightarrow R}). The iterative method is obtained by the Hermite inverse interpolation polynomial. We show that the convergence order of this method is given by the unique positive solution of a polynomial equation with coefficients given by the multiplicity orders. We consider the particular instance of two interpolation nodes and we determine the resulted methods.
Authors
Crăciun Iancu
(Tiberiu Popoviciu Institute of Numerical Analysis)
Ion Păvăloiu
(Tiberiu Popoviciu Institute of Numerical Analysis)
Title
Original title (in French)
La resolution des équations par interpolation inverse de type Hermite
English translation of the title
Solving equations by Hermite inverse interpolation
Keywords
Hermite interpolation; inverse interpolation; nonlinear equations in R; iterative methods; multistep method; convergence order
Cite this paper as:
C. Iancu, I. Păvăloiu, La resolution des équations par interpolation inverse de type Hermite, Mathematica (Cluj), 26(49) (1984) no. 2, pp. 115-123 (in French).
About this paper
Journal
Mathematica
Publisher Name
DOI
Not available yet.
Print ISSN
Not available yet.
Online ISSN
Not available yet.
References
[1] I.G. Berezin, P.N. Zidkov, Metody vycislenii, I. Moskow, 1962.
[2] Gh. Coman, Some practical approximation methods for nonlinear equations, Anal. Numer. Theor. Approx., 11, 1-2, (1982), 41–48.
[3] M.A. Ostrowski, Solution of Equations and Systems of Equations, Academic Press New York – London, 1960.
[4] I. Pavaloiu, Rezolvarea ecua¸iilor prin interpolare, Ed. Dacia, 1981.
[5] I. Pavaloiu, Introducere in aproximarea solutiilor ecuatiilor, Ed. Dacia, 1976.
[6] D.D. Stancu, Asupra formulei de interpolare a lui Hermite si a unor aplicatii ale acesteia, Studii ¸si Cercet. Mat. (Cluj), 3-4 VIII, (1957), 339-355.
[7] J.F. Traub, Iterative Methods for the Solution of Equations, Prentice Hall Series in Automatic Computation 1964.
[8] B.A. Turowicz, Sur les derivees d’ordre superiour d’une fonction inverse, Colloq. Math., (1959), 83–87.
Paper (preprint) in HTML form
Solving equations by Hermite type inverse interpolation
1.
In this work we study some numerical methods for solving equations of the form:
| (1) |
Oris a real function of a real variable, andis an interval of the real axis.
We designate bya set ofdistinct real numbers of the setthat's to say,Or For
Let us designate by
| (2) |
natural numbers that satisfy the condition:
| (3) |
Or, and by
| (4) |
given real numbers.
The following theorem is well known.
Theorem 1 .
Thehas a single polynomialof at least degreewhich verifies the following conditions:
| (5) |
The polynomialhas the following form:
| (6) |
Or
| (7) |
If we assume that the functionadmits derivatives up to orderinclusive in the intervaland ifthen the remainder of the Hermite interpolation formula satisfies the following equality:
| (8) |
Or
Let's suppose thatfor eachand designate bythe image of the intervalby functionFrom the above hypothesis it follows that the function is bijective, that the function admits derivatives up to orderinclusive in the intervalThe order derivative at one pointhas the following form [ 4 ] , [ 8 ] :
| (9) |
where the above sum extends to all integer, non-negative solutions of the system of equations:
| (10) |
If we assume that equation ( 1 ) has a root then ofit follows thatand offor eachit follows that is unique.
If in the expression of the Hermite interpolation polynomial ( 6 ) we choose for the interpolation nodes the valuesand for the values of the function and its derivatives the numbersrespectivelythen we obtain the Hermite interpolation polynomial of the functionThis polynomial will have the following form:
| (11) |
Or
| (12) |
The polynomial defined by the relation ( 11 ) is called the inverse Hermite interpolation polynomial.
From equality ( 8 ) it follows:
| (13) |
Taking into account that( 13 ) gives us
| (14) |
Oris contained in the smallest interval that contains the points:If we designate bythe interval above and bythe number then we deduce from ( 14 )
| (15) |
If the points of the setare close enough to, then the numbersare close to zero. Even more so the numberis close to zero and the choice ofas an approximation for the rootof equation ( 1 ) is justified. Therefore
| (16) |
2. Iterative methods obtained using the inverse Hermite interpolation polynomial
Si l’approximation ofgiven by ( 16 ) is not suitable, then by successive iterations we can obtain in certain cases approximations closer and closer to the root.
Let us designate by
| (17) |
initial approximations of the rootof equation ( 1 ). Using these approximations we construct the sequence given by the following iterative process:
| (18) |
Orhas the expression given by ( 11 ) andhas the following expression:
| (19) | |||
or
| (20) |
In the following we study the convergence of the sequencegenerated by the iterative process ( 18 ). We denote by
then from ( 14 ) and ( 18 ) it results:
| (21) |
More generally if the elements of the sequenceare contained in the intervalwe have the following inequalities:
| (22) | ||||
By multiplying the terms of ( 22 ) byand we designate bythe number
the inequalities ( 22 ) become:
| (23) |
We assume in the following:
| (24) |
Oris the positive root of the equation
| (25) |
And
If we assume
| (26) |
then taking into account ( 23 ) and ( 25 ) we have
| (27) |
It follows from ( 24 ) that
| (28) |
It is easy to show that equation ( 25 ) has a single real and positive rootThen from ( 27 ), ( 26 ), ( 28 ) and taking into account thatis the root of the equation ( 25 ) we deduce the following equalities:
| (29) |
that's to say
| (30) |
From equality ( 3 ) it follows thatAnd Therefore
what needed to be demonstrated.
The expression ofprovides the following assessment:
| (31) |
3. The special case
Let us designate bytwo numbers ofsuch as , and bytwo points of the intervalSuppose that in each pointwe know the values of the function and its successive derivatives up to the orderrespectively.
Using formula ( 9 ) we can calculate the values of the functionin the pointsand the values of these successive derivatives up to the orderrespectively.
The inverse Hermite interpolation polynomial over knots will have the following form:
| (32) |
Or
| (33) |
Taking in ( 32 )we obtain a first approximation of the rootof equation ( 1 ). If we denote bythis approximation then we have:
| (34) | ||||
From ( 34 ) and ( 14 ) we deduce
| (35) |
Using equality
| (36) |
where bywe denote the divided difference of the functionon the knotswe deduce from ( 35 ) the following inequality:
| (37) |
Or
If we consider the numbersAndas initial approximations for the rootfrom equation ( 1 ), we obtain from ( 19 ) fora sequel, with
| (38) | |||
| (39) |
Assuming that all elements of the sequence are contained in the interval we have the following inequalities:
| (40) |
If we multiply the inequalities ( 40 ) byand if we take into account thatwe get:
| (41) |
In the following we designate bythe names
| (42) |
Then the inequalities ( 41 ) become:
| (43) |
We will assume that
| (44) |
OrAndis the positive root of the equation
| (45) |
From ( 43 ) we deduce:
| (46) |
where the sequelchecks for equalities
| (47) | ||||
From ( 45 ) we deduce:
| (48) |
in which case ( 47 ) gives
| (49) |
Deduced forthe following assessments:
| (50) |
The previous inequality and ( 42 ) imply:
| (51) |
By designating bythe number
| (52) |
and assumingwe deduce from ( 51 )
| (53) |
which represents an evaluation of the error afterno iteration.
Passing to limit whenwe deduce from ( 53 )
| (54) |
and from ( 51 ) it results, from the fact thatis a continuous function,
In the following we will present an analysis of the convergence speed of the sequencegiven by ( 38 ). To do this we note that if we consider in the Hermite interpolation polynomial the knothaving the order of multiplicityand the knothaving the order of multiplicitythe equation which gives us the speed of convergence of the sequence( 45 ) will take the form:
| (55) |
It can be easily demonstrated that, ifand is the positive root of equation ( 55 ), then:
| (56) |
From the above considerations we conclude that the maximum order of convergence is obtained when in the two-node Hermite type inverse interpolation the multiplicity order ofis at least equal to the order of multiplicity ofIt goes without saying that in the above conclusion we assumed that the elements of the sequenceare obtained using the relation ( 38 ).
In the following we present two special cases of the Hermite inverse interpolation polynomial and the related iterative methods.
1.
In this case the inverse Hermite interpolation polynomial has the following form:
| (57) |
Taking into account thatAndand doing in ( 57 )we obtain the first approximation ofgiven by the relation:
| (58) |
Proceeding step by step we obtain the following:given by the relation:
| (59) |
which represents the rope method. In this case the followinghas the order of convergence
2. [2].
The inverse Hermite interpolation polynomial in this case takes the following form:
| (60) | ||||
Taking into account equalitiesAnd and by doing in ( 60 )we deduce:
| (61) |
Proceeding step by step we obtain the following:
| (62) | ||||
Or
The order of convergence of this method is
Bibliography
- [1] I.G. Berezin, P.N. Zidkov, Computation Methods , I. Moscow, 1962.
- [2] Gh. Coman, ††margin: clickable Some practical approximation methods for nonlinear equations, Anal. Numér. Théor. Approx., 11, 1-2, (1982), 41–48.
- [3] M.A. Ostrowski, Solution of Equations and Systems of Equations, Academic Press New York - London, 1960.
-
[4]
I. Pavaloiu,
††margin:
clickable
clickable
clickable Solving equations by interpolation , Ed. Dacia, 1981. - [5] I. Păvăloiu, Introduction to the approximation of solutions of equations , Ed. Dacia, 1976.
- [6] DD Stancu, On Hermite's interpolation formula and some of its applications , Studii si Cercet. Mat. (Cluj), 3-4 VIII, (1957), 339-355.
- [7] J.F. Traub, Iterative Methods for the Solution of Equations, Prentice Hall Series in Automatic Computation 1964.
- [8] BA Turowicz, On the higher order derivatives of an inverse function , Colloq. Math., (1959), 83–87.
Received on 06.06.1982
Institute of Computing
3400 Cluj-Napoca, Romania
