• Nu S-Au Găsit Rezultate

View of Double inequalities for quadrature formula of Gauss type with two nodes

N/A
N/A
Protected

Academic year: 2022

Share "View of Double inequalities for quadrature formula of Gauss type with two nodes"

Copied!
5
0
0

Text complet

(1)

Rev. Anal. Num´er. Th´eor. Approx., vol. 37 (2008) no. 1, pp. 53–57 ictp.acad.ro/jnaat

DOUBLE INEQUALITIES FOR QUADRATURE FORMULA OF GAUSS TYPE WITH TWO NODES

MARIUS HELJIU

Abstract. In this paper upper and lower error bonds for Gauss’s quadrature rule with two nodes are given.

MSC 2000. 65D32.

Keywords. Double integral inequalities, numerical integration.

1. INTRODUCTION

In a series of papers (see [2], [3], [8]) the authors establish bounds for the quadrature rules such as the trapezoid, Simpson and Newton quadrature rules.

In this work we will consider Gauss’s quadrature rule with two nodes:

(1)

Z b a

f(x)dx= ba

2 [f(x1) +f(x2)] +R[f]

where f : [a, b] → R, x1 = a+b2b−a2 ·ξ, x2 = a+b2 + b−a2 ·ξ, ξ = 1

3 = 0,57735027. . .

IffC4[a, b], the error R[f] from the formula (1) is given by:

(2) R[f] =

Z b a

ϕ(x)f(4)(x)dx

(see [7], pp. 137–138 and 283–284), where the function ϕhas the form:

(3) ϕ(x) =

(x−a)4

4! if x∈[a, x1],

(x−a)4

4!b−a2 (x−x3!1)3 if x∈]x1, x2[,

(b−x)4

4! if x∈[x2, b].

It is easy to see that the functionϕ has the following properties:

a) ϕC4[a, b];

b) ϕa+b2h=ϕa+b2 +h, for anyhh0,b−a2 i;

Department of Mathematics, University of Petro¸sani, Romania, e-mail:

[email protected].

(2)

c) Z b

a

ϕ(x)dx= 1351 b−a2 5.

2. MAIN RESULTS

In the following theorem double inequalities for−R[f] are presented where R[f] is the error in the quadrature formula (1) given by the relation (2).

Theorem 1. If fC4[a, b] then

1

17280(b−a)5(41γ4−45S3+ 180ξ3S3−180ξ3γ4)≤ (4)

b−a2 [f(x1) +f(x2)]− Z b

a

f(x)dx

172801 (b−a)5(41Γ4−45S3+ 180ξ3S3−180ξ3Γ4)

where γ4,Γ4 ∈ R, γ4f(4)(x) ≤Γ4, for all x∈ [a, b] and S3 = f000(b)−fb−a000(a). Moreover,

(5) γ4 = min

x∈[a,b]f(4)(x), Γ4= max

x∈[a,b]f(4)(x), and the inequalities (4) are sharp.

Proof. From (2), using the properties of the function ϕand integrating by parts, we obtain:

(6)

Z b a

ϕ(x)f(4)(x)dx= Z b

a

f(x)dx−b−a2 [f(x1) +f(x2)].

By using the equality c) in the formula (6) and the assumptions of the theorem, we have:

Z b a

[f(4)(x)−γ4]ϕ(x)dx= Z b

a

f(x)dx−b−a2 [f(x1) +f(x2)]

(7)

1351 b−a2 5γ4 and

Z b a

4f(4)(x)]ϕ(x)dx=− Z b

a

f(x)dx+b−a2 [f(x1) +f(x2)]

(8)

+1351 b−a2 5Γ4 On the other hand, we have:

(9)

Z b a

[f(4)(x)−γ4]ϕ(x)dx≤ max

x∈[a,b]|ϕ(x)|

Z b a

|f(4)(x)−γ4|dx, where

(10) max

x∈[a,b]|ϕ(x)|= 3841 (1−4ξ3)(b−a)4= 9−4

3

3456 (b−a)4,

(3)

and

Z b a

|f(4)(x)−γ4|dx= Z b

a

(f(4)(x)−γ4)dx (11)

=f000(b)−f000(a)−γ4(b−a)

= (S3γ4)(b−a).

From the relations (7), (9), (10) and (11) we obtain:

Z b a

f(x)dx−b−a2 [f(x1) +f(x2)]

(12)

≤ −172801 (b−a)5(41γ4−45S3+ 180ξ3S3−180ξ3γ4).

In the same way we have:

(13)

Z b a

4f(4)(x)]ϕ(x)dx≤ max

x∈[a,b]|ϕ(x)|

Z b a

4f(4)(x)|dx and

Z b a

4f(4)(x)|dx= Z b

a

4f(4)(x))dx (14)

= Γ4(b−a)f000(b) +f000(a)

= (Γ4S3)(b−a).

Using (8), (10), (13) and (14) we obtain the inequality:

Z b a

f(x)dx−b−a2 [f(x1) +f(x2)]

(15)

≥ −172801 (b−a)5(41Γ4−45S3+ 180ξ3S3−180ξ3γ4).

Inequalities (4) follow from the inequalities (12) and (15).

To prove the second part of the theorem we consider the function f(x) = (x−a)4. It is easy to show that all the three members of the double inequality (4) have in common the value−1801 (b−a)5. This completes the proof.

In the following theorem double inequalities forR[f] are presented.

Theorem 2. If the functionfC4[a, b]then:

1

17280(b−a)5(49γ4−45S3+ 180ξ3S3−180ξ3γ4) (16)

Z b

a

f(x)dx−ba

2 [f(x1) +f(x2)]

172801 (b−a)5(49Γ4−45S3+ 180ξ3S3−180ξ3Γ4), where γ4,Γ4, ξ and S3 are given in Theorem 1. Moreover,

γ4 = min

x∈[a,b]f(4)(x), Γ4= max

x∈[a,b]f(4)(x), and the inequalities (16) are sharp.

(4)

Proof. By using the relations (7), (9), (10) and (11) it follows:

Z b

a

f(x)dx+b−a2 [f(x1) +f(x2)]

(17)

≤ −172801 (b−a)5(49γ4−45S3+ 180ξ3S3−180ξ3γ4).

Analogously, using the relations (8), (13), (14) and (15) we obtain:

Z b a

f(x)dxb−a2 [f(x1) +f(x2)]

(18)

172801 (b−a)5(49Γ4−45S3+ 180ξ3S3−180ξ3Γ4).

From the relations (17) and (18) result the inequalities (16). To prove that the double inequalities (16) are exact we follow the steps of the proof for

Theorem 1.

Theorem 3 gives us the inequalities which do not depend onS3. Theorem 3. In the assumptions of Theorem1, we have:

1

34560(b−a)5(41γ4−49Γ4−180ξ3γ4+ 180ξ3Γ4) (19)

b−a2 [f(x1) +f(x2)]− Z b

a

f(x)dx

345601 (b−a)5(41Γ4−49γ4+ 180ξ3γ4−180ξ3Γ4).

If

γ4 = min

x∈[a,b]f(4)(x), Γ4= max

x∈[a,b]f(4)(x), then the inequalities (19) are sharp.

Proof. Multiplying the inequality (16) with (−1) and adding it with the inequality (4) we obtain the double inequalities (19). Considering the function f(x) = (x−a)4 and calculating all the three members of the inequality (19) the value obtained is −1801 (b−a)5. The double inequalities are sharp. This

completes the proof.

REFERENCES

[1] Cerone, P., Three Points Rules in Numerical Integration, Nonlinear Anal. Theory Methods Appl.,47, no. 4, pp. 2341–2352, 2001.

[2] Cerone, P. andDragomir, S.S.,Midpoint-Type Rules from an Inequalities Point of View, Handbook of Analytic-Computational Methods in Applied Mathematics, Editor:

G. Anastassiou, CRC Press, New York, pp. 135–200, 2000.

[3] Cerone, P.andDragomir, S.S.,Trapezoidal-Type Rules from an Inequalities Point of View, Handbook of Analytic-Computational Methods in Applied Mathematics, Editor:

G. Anastassiou, CRC Press, New York, pp. 65–134, 2000.

[4] Dragomir, S.S., Agarwal, R.P. andCerone, P.,On Simpson’s Inequality and Ap- plications, J. Inequal. Appl.,5, pp. 533–579, 2000.

[5] Dragomir, S.S., Pecari´c, J. and Wang, S., The Unified Treatment of Trapezoid, Simpson and Ostrowski Type Inequalities for Monotonic Mappings and Applications, Math. Comput. Modelling, 31, pp. 61–70, 2000.

(5)

[6] Heljiu, M.,Double Inequalities of Newton’s Quadrature Rule, Revue D’Analyse Nu- merique et de Theorie de L’Aproximation,35, no. 2, pp. 141–147, 2006.

[7] Ionescu, D.V.,Numerical Integration, Ed. Tehnic˘a, Bucure¸sti, 1957 (in Romanian).

[8] Ujevic, N., Some Double Integral Inequalities and Applications, Acta Math. Univ.

Comenianae,71, no. 2, pp. 187, 2002.

[9] Ujevic, N., Double Integral Inequalities of Simpson Type and Applications, J. Appl.

Math. & Computing,14, no. 1-2, pp. 213–223, 2004.

Received by the editors: March 11, 2008.

Referințe

DOCUMENTE SIMILARE

It is easy to see that we get the same result for the remain- der term of the composite Bernstein type quadrature formula as the result obtained by D.. Stancu in [7], for the

S., Midpoint-type Rules from an Inequalities Point of View, Handbook of Analytic-Computational Methods in Applied Mathematics, Editor:.. Anastassiou, CRC Press, New

(1 − α)C, with α ∈ (0, 1), G, C being the geometric and anti-harmonic means, and we find the range of values of α for which the weighted mean is still greater or less than

Goldman has defined in [3] an approximation operator of Stancu type for which has proved that it preserves the first degree functions.. In this paper, we begin the study of the

Before we prove the second result, we need the following lemma providing an identity that is interesting in itself as well..

In the present paper we consider weighted integrals and develop ex- plicit quadrature formulae of Gauss–Christoffel–Stancu type using simple Gauss- ian nodes and multiple fixed

Ky Fan studied in [4] the existence of solutions for some systems of convex inequalities involving lower semicontinuous functions defined on a compact convex set in a topological

'I'he autho¡s giyc árr eirsy coustluctive lrloccss and LhcS' pl'o\¡e some ll.tot.tototl¡' properties al1cl.. cortvexiLy