• Nem Talált Eredményt

Some properties of monotonicity and convexity of the q-Schurer -Stancu operators are considered

N/A
N/A
Protected

Academic year: 2022

Ossza meg "Some properties of monotonicity and convexity of the q-Schurer -Stancu operators are considered"

Copied!
10
0
0

Teljes szövegt

(1)

Miskolc Mathematical Notes HU e-ISSN 1787-2413 Vol. 19 (2018), No. 1, pp. 19–28 DOI: 10.18514/MMN.2018.1785

ON THE MONOTONICITY OFq-SCHURER-STANCU TYPE POLYNOMIALS

ANA MARIA ACU, CARMEN VIOLETA MURARU, AND VOICHIT¸ A ADRIANA RADU

Received 05 October, 2015

Abstract. Some properties of monotonicity and convexity of the q-Schurer

-Stancu operators are considered. The paper contains also numerical examples based on Matlab algorithms, which verify these properties.

2010Mathematics Subject Classification: 41A10; 41A36

Keywords: generalized Schurer-Stancu operators, q-integers, monotonicity, convexity

1. PRELIMINARIES

In the last decades, the application of q-calculus represents one of the most in- teresting areas of research in approximation theory. Lupas¸ [12] introduced in 1987 a q-type of the Bernstein operators and in 1997 another generalization of these op- erators based on q-integers was introduced by Phillips [16]. Their approximation properties were studied by Videnskiˇi [18], N. Mahmudov [13], T. Acar and A. Aral [1] and O. Dalmanoglu [9,10]. In time, many authors have been studied new classes of q-generalized operators ([2–4,6,7,17]).

Before proceeding further, we mention some basic definitions and notations from q-calculus. For any fixed real numberq > 0, the q-integerŒkq, fork2Nis defined as

ŒkqD 8

<

: 1 qk

1 q ; q¤1;

k; qD1:

The q-factorial integer and the q-binomial coefficients are : ŒkqŠD

ŒkqŒk 1q: : : Œ1q; kD1; 2; : : :

1; kD0;

n k

qD ŒnqŠ

ŒkqŠŒn kqŠ; .nk0/:

c 2018 Miskolc University Press

(2)

The q-analoque of.x a/nqis the polynomial .x a/nqD

1; ifnD0;

.x a/.x qa/ : : : .x qn 1a/; ifn1:

Let p be a non-negative integer and let ˛; ˇ be some real parameters satisfying the conditions 0˛ ˇ. In 2003, D. B˘arbosu [8] introduced for any f 2C Œ0; 1Cpandx2Œ0; 1the Schurer-Stancu operators as follows

Sm;p.˛;ˇ /.f; q; x/D

mCp

X

kD0

pm;k.x/f

kC˛ mCˇ

;

wherepm;k.x/D

mCp k

xk.1 x/mCp k.

Recently, P.N. Agrawal, V. Gupta and A.S. Kumar [5] introduced the class of q- Schurer-Stancu operators. For anym2N,pa fixed non negative integer number and

˛; ˇsome real parameters satisfying the conditions0˛ˇ, they constructed the class of generalized q-Schurer-Stancu operators

SQm;p.˛;ˇ /WC Œ0; 1Cp!C Œ0; 1;

as follows

SQm;p.˛;ˇ /.f; q; x/D

mCp

X

kD0

Q

pm;k.x/f

ŒkqC˛ Œmq

; x2Œ0; 1; (1.1)

wherepQm;k.x/D

mCp k

q

xk.1 x/mqCp k.

If˛DˇD0the above operators reduce to the Bernstein-Schurer operators intro- duced by Muraru in [14].

Lemma 1([5]). For the operators defined in (1.1) the following properties hold 1.SQm;p.˛;ˇ /.e0; q; x/D1;

2.SQm;p.˛;ˇ /.e1; q; x/D ˛

ŒmqCˇCŒmCpq ŒmqCˇx;

3.

SQm;p.˛;ˇ /.e2; q; x/

D ˛2

.ŒmqCˇ/2C ŒmCp2q

.ŒmqCˇ/2x2C2˛ŒmCpqx

.ŒmqCˇ/2 CŒmCpqx.1 x/

.ŒmqCˇ/2 : The next result is based on Popoviciu’s technique and it is expressed in terms of the first order modulus of continuity.

(3)

ON THE MONOTONICITY OF 21

Theorem 1([5]). Iff 2C Œ0; 1Cpandq2.0; 1/then

SQm;p.˛;ˇ /.f; q; x/ f .x/

5

4!fm/ (1.2)

holds, where

ımD 1 Œmq

q

ŒmCpqC4.qmŒpqC˛ ˇ/2:

2. MONOTONICITY OF THEq-SCHURER-STANCU OPERATORS

Oruc and Philips [15] showed that for a convex functionf on [0,1], the q-Bernstein polynomials are monotonic decreasing. In this section we will prove a similar result forq-Schurer-Stancu operators.

Theorem 2. Letf be a convex and increasing function onŒ0; pC1. Then, for 0 < q1andˇŒpq

qp ,

SQm 1;p.˛;ˇ / .f; q; x/ QSm;p.˛;ˇ /.f; q; x/; (2.1) for0x1andm2.

Proof. For0 < q < 1we have

mCp 1

Y

sD0

.1 qsx/ 1h

SQm 1;p.˛;ˇ / .f; q; x/ SQm;p.˛;ˇ /.f; q; x/i

D

mCp 1

X

kD0

mCp 1 k

q

xk

mCp 1

Y

sDmCp k 1

.1 qsx/ 1f

ŒkqC˛ Œm 1q

mCp

X

kD0

mCp k

q

xk

mCp 1

Y

sDmCp k

.1 qsx/ 1f

ŒkqC˛ Œmq

:

Denote

k.x/Dxk

mCp 1

Y

sDmCp k

.1 qsx/ 1 (2.2)

and using the following relation xk

mCp 1

Y

sDmCp k 1

.1 qsx/ 1D k.x/CqmCp k 1 kC1.x/

we find

mCp 1

Y

sD0

.1 qsx/ 1h

SQm 1;p.˛;ˇ / .f; q; x/ SQm;p.˛;ˇ /.f; q; x/i

(4)

D

mCp 1

X

kD0

f

ŒkqC˛ Œm qCˇ

mCp 1 k

q

n

k.x/CqmCp k 1 kC1.x/o

mCp

X

kD0

f

ŒkqC˛ ŒmqCˇ

mCp k

q

k.x/D

mCp 1

X

kD0

f

ŒkqC˛ Œm qCˇ

mCp 1 k

q k.x/

C

mCp

X

kD1

qmCp kf

Œk qC˛ Œm qCˇ

mCp 1 k 1

q k.x/

mCp

X

kD0

f

Œkq

ŒmqCˇ

mCp k

q

k.x/D

mCp 1

X

kD1

( f

Œkq

Œm qCˇ

mCp 1 k

q

CqmCp kf

Œk qC˛ Œm qCˇ

mCp 1 k 1

q

f

ŒkqC˛ ŒmqCˇ

mCp k

q

)

k.x/

C

f

ŒmCp qC˛ Œm qCˇ

f

ŒmCqC˛ ŒmqCˇ

mCp.x/

C

f

˛ Œm qCˇ

f

˛ ŒmqCˇ

0.x/

D

mCp 1

X

kD1

mCp k

q

ak k.x/C

f

ŒmCp 1qC˛ Œm qCˇ

f

ŒmCqC˛ ŒmqCˇ

mCp.x/

C

f

˛ Œm qCˇ

f

˛ ŒmqCˇ

0.x/;

where ak Df

ŒkqC˛ Œm 1q

ŒmCp kq

ŒmCpq CqmCp kf

Œk 1qC˛ Œm 1q

Œkq

ŒmCpq

f

ŒkqC˛ Œmq

:

From (2.2) it is clear that each k.x/is non-negative on Œ0; 1for0q1and thus, it suffices to show that eachak is non-negative.

Sincef is convex on Œ0; pC1, for anyt0; t1 such that0t0< t1pC1and any,0 < < 1, we have

f .t0C.1 /t1/f .t0/C.1 /f .t1/: (2.3) Let t0 D Œk 1q

Œm 1qCˇ, t1 D Œkq

Œm 1qCˇ and DqmCp k Œkq

ŒmCpq

. Then 0t0< t1pC1and0 < < 1 for1kmCp 1. If we replace them in the relation (2.3), it follows

qmCp k Œkq

ŒmCpq

f

Œk 1qC˛ Œm 1q

CŒmCp kq

ŒmCpq

f

ŒkqC˛ Œm 1q

(5)

ON THE MONOTONICITY OF 23

f

qmCp k Œkq

ŒmCpq Œk 1q

Œm 1qCˇCŒmCp kq

ŒmCpq ŒkqC˛ Œm 1q

: Using the inequalityŒkq.Œk 1qC˛/.ŒkqC˛/Œk 1qandf increasing func- tion, it follows

qmCp k Œkq

ŒmCpq

f

Œk 1qC˛ Œm 1q

CŒmCp kq

ŒmCpq

f

ŒkqC˛ Œm 1q

(2.4)

f Œkq

ŒmCpq qmCp kŒk 1qCŒmCp kq

Œm 1q

! Df

Œkq

ŒmCpqŒmCp 1q

Œm 1q

:

Sincef is increasing onŒ0; pC1and akDf

ŒkqC˛ Œm 1q

ŒmCp kq

ŒmCpq CqmCp kf

Œk 1qC˛ Œm 1q

Œkq

ŒmCpq

f

Œkq

ŒmCpq ŒmCp 1q

Œm 1q

C

f

Œkq

ŒmCpq ŒmCp 1q

Œm 1q

f

ŒkqC˛ Œmq

;

from the inequality (2.4) we obtainak0,kD1; mCp 1.

ThereforeSQm 1;p.˛;ˇ / .f; q; x/ QSm;p.˛;ˇ /.f; q; x/:

ForqD1and0x < 1in a similar way the property (2.1) is verified.

ForqD1andxD1we have Q

Sm 1;p.˛;ˇ / .f; 1; 1/ SQm;p.˛;ˇ /.f; 1; 1/Df

mCp 1C˛ m 1Cˇ

f

mCpC˛ mCˇ

0:

Theorem 3. Iff is convex, then for allm1and0 < q1it follows

i) SQm;p.˛;ˇ /.f; q; x/ f .x/, for x 2 Œ0; 1, f increasing on Œ0; 1 and ˇD˛C","2

0; qmŒpq

; ii) SQm;p.˛;ˇ /.f; q; x/f .x/, forx2

0; ˛

ˇ qmŒpq

,f increasing onŒ0; 1and ˇ > ˛CqmŒpq;

iii) SQm;p.˛;ˇ /.f; q; x/f .x/, forx2

˛ ˇ qmŒpq

; 1

,f decreasing onŒ0; 1and ˇ > ˛CqmŒpq.

Proof. We consider the knotesxkD Œkq

ŒmqCˇ,0kmCp. From Lemma1it follows

mCp

X

kD0

Q

pm;k.x/D1;

mCp

X

kD0

Q

pm;k.x/xk D ˛

ŒmqCˇCŒmCpq

ŒmqCˇx:

(6)

Using the convexity of function f we have SQm;p.˛;ˇ /.f; q; x/D

mCp

X

kD0

Q

pm;k.x/f .xk/f

mCp

X

kD0

Q

pm;k.x/xk

!

Df

˛

ŒmqCˇCŒmCpq

ŒmqCˇx

:

The following inequalities hold

a) ˛

ŒmqCˇCŒmCpq

ŒmqCˇxxforˇD˛C,2Œ0; qmŒp,x2Œ0; 1;

b) ˛

ŒmqCˇCŒmCpq

ŒmqCˇxxforˇ > ˛CqmŒpq,x2h

0;ˇ qm˛Œpq

i

;

c) ˛

ŒmqCˇCŒmCpq

ŒmqCˇxxforˇ > ˛CqmŒpq,x2h

˛

ˇ qmŒpq; 1i . The theorem is proved using the monotony of function f and the inequalities a)-

c).

3. NUMERICAL EXAMPLE

Davis [11] proved that for any convex functionf, the classical Bernstein polyno- mial is convex and the sequence of Bernstein polynomials is monotonic decreasing.

Oruc and Philips [15] extend these results for the Bernstein operators in q-calculus for 0 < q 1 . In this section we will verify numerically these properties for the q-Schurer-Stancu operators.

TABLE1. The q-Schurer-Stancu operators x SQ30;6.3;5/.f; q; x/ SQ90;6.3;5/.f; q; x/

0 0:029115231597413 0:026565252311249 0:1 0:089811163243826 0:083383934253406 0:2 0:191687642176340 0:179475155093384 0:3 0:347058215579779 0:326749960898829 0:4 0:570050753689721 0:538887425606676 0:5 0:876786656660820 0:831510780091276 0:6 1:285573442130916 1:222376845885256 0:7 1:817111513724159 1:731579581814080 0:8 2:494715950321931 2:381768592612226 0:9 3:344554197167635 3:198383491142964 1 4:395900582819147 4:209905050209471

(7)

ON THE MONOTONICITY OF 25

In Table1are calculated the values of the q-Schurer-Stancu operatorsSQ30;6.3;5/.f; q; x/

andSQ90;6.3;5/.f; q; x/forf .x/Dx3exC1 andqD0:9. Also, in the Figure1are given the graphics of these operators.

0 0.2 0.4 0.6 0.8 1

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

S30,6(3,5) S90,6(3,5)

FIGURE1. The monotonicity of theq-Schurer-Stancu operators

In the next part of this section we will give some numerical examples which verify the inequalities proved in Theorem3.

Example 1. IfnD50, pD5, ˛D3, ˇD3:0211, qD0:9, f .x/Dx3exC1, it followsSQm;p.˛;ˇ /.f; q; x/f .x/for allx2Œ0; 1.

0 0.2 0.4 0.6 0.8 1

0 1 2 3 4 5 6 7 8

q−Schurer Stancu operator aproximated function

FIGURE2. Theq-Schurer-Stancu operators for increasing function andˇD˛C

(8)

Example 2. IfnD50, pD5, ˛D3, qD0:9, ˇD8:0211, f .x/Dx3exC1, it followsSQm;p.˛;ˇ /.f; q; x/f .x/for allx2Œ0; 0:375.

0 0.2 0.4 0.6 0.8 1

0 1 2 3 4 5 6 7 8

q−Schurer Stancu operator aproximated function

FIGURE3. Theq-Schurer-Stancu operators for increasing function andˇ > ˛CqmŒpq

Example3. IfnD50,pD5,˛D3,qD0:9,ˇD8:0211,f .x/De x2, it follows SQm;p.˛;ˇ /.f; q; x/f .x/for allx2Œ0:375; 1:

0 0.2 0.4 0.6 0.8 1

0.4 0.5 0.6 0.7 0.8 0.9 1

q−Schurer Stancu operator aproximated function

FIGURE 4. Theq-Schurer-Stancu operators for decreasing function

ACKNOWLEDGMENT.

Project financed from Lucian Blaga University of Sibiu research grants LBUS- IRG-2015-01, No.2032/7.

(9)

ON THE MONOTONICITY OF 27

REFERENCES

[1] T. Acar and A. Aral, “On pointwise convergence of q-bernstein operators and their q-derivatives,”

Numer. Funct. Anal. Optim., vol. 36, no. 3, pp. 287–304, 2015.

[2] A. Acu, “Stancu-schurer-kantorovich operators based on q-integers,”Appl. Math. Comput., vol.

259, pp. 896–907, 2015.

[3] O. Agratini, “On a q-analogue of stancu operators,”Central European Journal of Mathematics, vol. 8, no. 1, pp. 191–198, 2010.

[4] O. Agratini and C. Radu, “An extension based on qr-integral for a sequence of operators,”Appl.

Math. Comput., vol. 218, pp. 140–147, 2011.

[5] P. Agrawal, V. Gupta, and A. Kumar, “On q analogue of bernstein-schurer-stancu operators,”Appl.

Math. Comput., vol. 219, pp. 7754–7764, 2013.

[6] A. Aral and T. Acar, “Voronovskaya type result for q-derivative of q-baskakov operators,”J. Appl.

Funct. Anal., vol. 7, no. 4, pp. 321–331, 2012.

[7] A. Aral, V. Gupta, and R. P. Agarwal,Applications of q Calculus in Operator Theory. Springer, 2013.

[8] D. Barbosu, “Scurer-stancu type operators,”Stud. Univ. Babes¸-Bolyai Math., vol. XLVIII, no. 3, pp. 31–35, 2003.

[9] O. Dalmanoglu, “Approximation by kantorovich type q-bernstein operators,” in12th WSEAS Int.

Conf. on Applied Mathematics, Cairo, Egypte, 2007, pp. 29–31.

[10] O. Dalmanoglu and O. Dogru, “On statistical approximation properties of kantorovich type q- bernstein operators,”Mathematical and Computer Modelling, vol. 52, pp. 760–771, 2010.

[11] P. Davis,Interpolation and Approximation. New York: Dover Publications Inc., 1976.

[12] A. Lupas, “A q-analogue of the bernstein operator,”Seminar on Numerical and Statistical Calcu- lus, University of Cluj-Napoca, vol. 9, pp. 85–92, 1987.

[13] N. Mahmudov, “The moments for q-bernstein operators in the case0 < q < 1,”Numer. Algorithms, vol. 53, pp. 439–450, 2010.

[14] C. Muraru, “Note on q-bernstein-schurer operators,” Stud. Univ. Babes¸-Bolyai Math., vol. 56, no. 2, pp. 489–495, 2011.

[15] H. Oruc and G. Philips, “A generalization of the bernstein polynomials,”Proc. Edinb. Math. Soc., vol. 42, pp. 403–413, 1999.

[16] G. Philips, “On generalized bernstein polynomials,” inNumerical Analysis, D. Griffits and G. Wat- son, Eds., 1996, pp. 263–269.

[17] D. Sofonea, “On a q-analogue of a. lupas¸ operators,” inMathematical Analysis and Approxima- tion Theory, Proceedings of RoGer 2004, B˘ais¸oara, I. Gavrea and M. Ivan, Eds. Cluj-Napoca:

Mediamira Science Publisher, 2005, pp. 211–223.

[18] V. Videnskiˇi,Linear positive operators of finite rank. Leningrad: A. I. Gerzen State Pedagocical Institute, 1985.

(10)

Authors’ addresses

Ana Maria Acu

Lucian Blaga University of Sibiu, Department of Mathematics and Informatics, Str. Dr. I. Ratiu, No.5-7, RO-550012 Sibiu, Romania

E-mail address:acuana77@yahoo.com

Carmen Violeta Muraru

Vasile Alecsandri University of Bac˘au, Department of Mathematics, Informatics and Educational Sciences, Calea M˘ar˘as¸es¸ti 158, RO-600115 Bac˘au, Romania

E-mail address:carmen 7419@yahoo.com, cmuraru@ub.ro

Voichit¸a Adriana Radu

Babes-Bolyai University, FSEGA, Department of Statistics-Forecasts-Mathematics, Str. Teodor Mi- hali, No.58-60, RO-400591 Cluj Napoca, Romania

E-mail address:voichita.radu@econ.ubbcluj.ro, voichita.radu@gmail.com

Hivatkozások

KAPCSOLÓDÓ DOKUMENTUMOK

The paper is composed as following: the statistical properties of PageRank and CheiRank are analyzed in Section 2, ranking of Wikipedia personalities and uni- versities are

The main aim of this paper is to give some new Turán-type inequalities for the q-polygamma and q-zeta [2] functions by using a q-analogue of the generalization of the

Kirov in [8] defined the new Bernstein polynomials for m-times differentiable functions and showed that these operators have better approximation properties than classical

Key words: q−integers, q−Meyer-König-Zeller Durrmeyer type operators, A-Statistical convergence, Weighted space, Weighted modulus of smoothness, Lipschitz class.. Abstract: We

Key words and phrases: q−integers, q−Meyer-König-Zeller Durrmeyer type operators, A-Statistical convergence, Weighted space, Weighted modulus of smoothness, Lipschitz class..

Class wF (p, r, q) Operators and Quasisimilarity Changsen Yang and Yuliang

J. Pure and Appl. The following monotonicity properties of weighted Stolarsky means have been established in [9]:. Property 1.1.. Convexity of Weighted

Approximation theory has been used in the theory of approximation of continuous functions by means of sequences of positive linear operators and still there remains a very active