R E S E A R C H
Open Access
(
p
,
q
)
-gamma operators which preserve
x
2
Wen-Tao Cheng
1*, Wen-Hui Zhang
2and Qing-Bo Cai
3*Correspondence:
chengwentao_0517@163.com 1School of Mathematics and
Computation Science, Anqing Normal University, Anhui, China Full list of author information is available at the end of the article
Abstract
In this paper, we introduce (p,q)-gamma operators which preservex2, we estimate the moments of these operators, and establish direct and local approximation theorems of these operators. Then two approximation theorems about Lipschitz functions are obtained. The estimates on the rate of convergence and some weighted approximation theorems of the operators are also obtained. Furthermore, the Voronovskaja-type asymptotic formula is also presented.
MSC: 41A10; 41A25; 41A36
Keywords: (p,q)-integers; (p,q)-gamma operators; Modulus of continuity; Rate of convergence
1 Introduction
With the rapid development of the approximation theory about the operators since the last century, lots of operators, such as Bernstein operators [4], Szász–Mirakjan operators [32,
37], Baskakov operators [3], Bleimann–Butzer–Hann operators [5], and Meyer–König– Zeller operators [31], have been proposed and constructed by several researchers due to Weierstrass and the important convergence theorem of Korovkin [26], see also [17]. In [23], Karsli considered gamma operators and studied the rate of convergence of these op-erators for the functions with derivative of bounded variation
Ln(f;x) =
(2n+ 3)!xn+3
n!(n+ 2)!
∞
0
tn
(x+t)2n+4f(t) dt, x> 0. (1)
In [25], Karsli and Ozarslan established some local and global approximation results for the operatorsLn.
In recent years, with the rapid development ofq-calculus [22], the study of new poly-nomials and operators constructed withq-integer has attracted more and more attention. Lupas first introducedq-Bernstein polynomials [27], and Phillips [36] proposed otherq -analogue of Bernstein polynomials. Later, many researchers have performed studies in this field, and theq-analogue of classical operators and modified operators, such asq-Szász– Mirakjan operators [28], q-Baskakov operators [13], q-Meyer–König–Zeller operators [12], q-Bleimann–Butzer–Hann operators [11] q-Phillips operators [29], q-Baskakov– Kantorovich operators [20],q-Baskakov–Durrmeyer operators [19],q-Szász-beta oper-ators [18], andq-Meyer–König–Zeller–Durrmeyer operators [15], has been constructed;
see also [2]. In [6], Cai and Zeng definedq-gamma operators
Gn,q(f;x) =
[2n+ 3]!(qn+32x)n+3qn(n2+1)
[n]q![n+ 2]q!
∞
0
tn
(qn+32x+t)2n+4
q
f(t) dqt, x> 0 (2)
and gave their approximation properties.
Then many operators have been constructed with two parameters (p,q)-integer based on post-quantum calculus ((p,q)-calculus) which has been used efficiently in many areas of sciences such as Lie group, different equations, hypergeometric series, physical sci-ences, and so on. Recently, approximation by sequences of linear positive operators has been transferred to operators with (p,q)-integer. Let us review some useful notations and definitions about (p,q)-calculus in [2,17,21].
Let 0 <q<p≤1. For each nonnegative integern, the (p,q)-integer [n]p,q, (p,q)-factorial
[n]p,q! are defined by
[n]p,q= pn–qn
p–q , n= 0, 1, 2, . . .
and
[n]p,q! =
⎧ ⎨ ⎩
[1]p,q[2]p,q· · ·[n]p,q, n≥1;
1, n= 0.
Further, the (p,q)-power basis is defined by
(x⊕y)np,q= (x+y)(px+qy)p2x+q2y· · ·pn–1x+qn–1y
and
(xy)np,q= (x–y)(px–qy)p2x–q2y· · ·pn–1x–qn–1y.
Letnbe a non-negative integer, the (p,q)-gamma function is defined as
Γp,q(n+ 1) =
(pq)n p,q
(p–q)n = [n]p,q!, 0 <q<p≤1.
Aral and Gupta [1] proposed a (p,q)-beta function of the second kind form,n∈Nas follows:
Bp,q(m,n) =
∞
0
xm–1
(1⊕px)m+n p,q
dp,qx
and gave the relation of the (p,q)-analogues of beta and gamma functions:
Bp,q(m,n) =
qΓp,q(m)Γp,q(n)
(pm+1qm–1)m2Γp,q(m+n)
As a special case, ifp=q= 1,B(m,n) =ΓΓ(m(m)Γ+n(n)). It is obvious that order is important for (p,q)-setting, which is the reason why a (p,q)-variant of beta function does not satisfy commutativity property, i.e.,Bp,q(m,n)=Bp,q(n,m).
LetCB[0,∞) be the space of all real-valued continuous bounded functionsf on the
in-terval [0,∞) endowed with the norm
f = sup x∈[0,∞)
f(x).
Letδ> 0 andC2
B[0,∞) ={g:g,g∈CB[0,∞)}, the followingK-functional is defined:
K(f;δ) = inf g∈C2B[0,∞)
f–g +δg.
Using DeVore–Lorentz theorem (see [10]), there exists a constantC> 0 such that
K(f;δ)≤Cω2(f;
√
δ), (3)
where
ω2(f;δ) = sup 0<|t|≤δ
sup x∈[0,∞)
f(x+ 2t) – 2f(x+t) +f(x)
is the second order modulus of smoothness off. Also, byω(f;δ) we denote the usual mod-ulus of continuity off ∈CB[0,∞) defined as
ω(f;δ) = sup
0<|t|≤δ
sup x∈[0,∞)
f(x+t) –f(x).
LetBx2[0,∞) denote the function space of all functionsf such that|f(x)| ≤Cf(1 +x2),
whereCf is a positive constant depending onf. ByCx2[0,∞) we denote the subspace of all
continuous functions in the function spaceBx2[0,∞). ByC0
x2[0,∞) we denote the subspace
of all functionsf ∈Cx2[0,∞) for whichlimx→∞|f(x)|
1+x2 is endowed with the norm
f x2= sup
x∈[0,∞)
|f(x)| 1 +x2.
Fora> 0, the modulus of continuity off on [0,a] is defined as follows:
ωa(f;δ) = sup |y–x|<δ
sup
0≤x,y≤a
f(y) –f(x).
As is known, iff is not uniformly continuous on [0,∞), we cannot getω(f;δ)→0 as
δ →0. In [38], Yuksel and Ispir defined the weighted modulus of continuityΩ(f;δ) =
sup0<h≤δ,x≥0|f(x+h)–f(x)|
1+(x+h)2 whilef ∈Cx02[0,∞) and proved the properties of monotone
increas-ing about Ω(f;δ) asδ> 0 and the inequalityΩ(f;λδ)≤(1 +λ)Ω(f;δ) whileλ> 0 and
f ∈Cx02[0,∞).
Letf ∈CB[0,∞),M> 0, andγ ∈(0, 1]. We recall thatf ∈LipM(γ) if the following
in-equality
is satisfied. LetF be a subset of the interval [0,∞), we define thatf ∈LipM(γ,F) if the following inequality
f(x) –f(y)≤M|x–y|γ, x∈Fandy∈[0,∞)
holds.
Recently, Mursaleen first applied (p,q)-calculus in approximation theory and introduced the (p,q)-analogue of Bernstein operators [33], (p,q)-Bernstein–Stancu operators [34], (p,q)-Bernstein–Schurer operators [35] and investigated their approximation properties. In addition, many well-known approximation operators with (p,q)-integer, such as (p,q )-Bernstein–Stancu–Schurer–Kantorovich operators [8], (p,q)-Szász–Baskakov operators [16], (p,q)-Baskakov-beta operators [30] have been introduced. All this achievement mo-tivates us to construct the (p,q)-analogue of the gamma operator (1), as we know that many researchers have studied approximation properties of the gamma operators and their modifications (see [7,9,24,39]). The rest of the paper is organized as follows. In Sect.2, we define the (p,q)-gamma operators and obtain the moments and the central moments of them. In Sect.3, we study the properties of the (p,q)-gamma operators about Lipschitz condition. Then some direct theorems about local approximation, rate of con-vergence, weighted approximation, and Voronovskaja-type approximation are obtained.
2 (p,q)-gamma operators and moments
We first define the analogue of gamma operators via (p,q)-calculus as follows.
Definition 2.1 For n∈N,x∈(0,∞)and0 <q<p≤1,the(p,q)-gamma operators can be defined as follows:
Gpn,q(f;x) =x
n+3(qn+32)n+3pn2+7 2n+72
Bp,q(n+ 1,n+ 3)
∞
0
tn
((pq)n+32x⊕t)2n+4
p,q
f(t) dp,qt.
OperatorsGpn,qare linear and positive. Forp= 1, they turn out to be theq-gamma
op-erators defined in (2). We will derive the momentsGpn,q(tk;x) and the central moments Gpn,q((t–x)k;x) fork= 0, 1, 2, 3, 4.
Lemma 2.1 For x∈(0,∞), 0 <q<p≤1,and k= 0, 1, . . . ,n+ 2,we have
Gpn,qtk;x=x
k(pq)k–k22[n+k]p,q![n–k+ 2]p,q!
[n]p,q![n+ 2]p,q!
. (4)
Proof Using the properties of (p,q)-beta function and (p,q)-gamma function, we have
Gpn,qtk;x=x
n+3(qn+32)n+3pn2+72n+72
Bp,q(n+ 1,n+ 3)
∞
0
tn+k
((pq)n+32x⊕t)2n+4
p,q
dp,qt
=x
n+3(qn+32)n+3pn2+72n+72
Bp,q(n+ 1,n+ 3)
∞
0
1 (pq)(2n+3)(n+2)x2n+4
× tn+k
(1⊕ pt
xqn+ 32pn+ 52
)2n+4
p,q
=x
n+3(qn+32)n+3pn2+72n+72
Bp,q(n+ 1,n+ 3)
∞
0
(xqn+32pn+52)n+k+1
(pq)(2n+3)(n+2)x2n+4
×
( t xqn+ 32pn+ 52
)n+k
(1⊕ pt
xqn+ 32pn+ 52
)2n+4
p,q
dp,q
t
xqn+32pn+52
=x
kpkn+52kqkn+32kB
p,q(n+k+ 1,n–k+ 3) Bp,q(n+ 1,n+ 3)
=x
k(pq)k–k22[n+k]p,q![n–k+ 2]p,q!
[n]p,q![n+ 2]p,q!
.
Lemma2.1is proved.
Lemma 2.2 For x∈(0,∞), 0 <q<p≤1,the following equalities hold:
1.Gpn,q(1;x) = 1;
2.Gpn,q(t;x) =
p q(1 –
pn+1
[n+2]p,q)x; 3.Gpn,q(t2;x) =x2;
4.Gpn,q(t3;x) = [n+3]p,qx
3
(pq)32[n]p,q ;
5.Gpn,q(t4;x) =
[n+3]p,q[n+4]p,qx4
(pq)4[n]
p,q[n–1]p,q for n> 1.
Proof The proof of this lemma is an immediate consequence of Lemma2.1. Hence the
details are omitted.
Lemma 2.3 Let n> 1and x∈(0,∞),then for0 <q<p≤1,we have the central moments
as follows:
1. A(x) :=Gnp,q(t–x;x) = ((
p q– 1) –
p q
pn+1
[n+2]p,q)x;
2. B(x) :=Gnp,q((t–x)2;x) = –2((
p q– 1) –
p q
pn+1
[n+2]p,q)x
2; 3. Gpn,q((t–x)4;x) = ([n+2]p,q[n+3]p,q[n+4]p,q–4(pq)
5
2[n–1]p,q[n+2]p,q[n+3]p,q
(pq)4[n–1]
p,q[n]p,q[n+2]p,q +
–4(pq)92[n–1]p,q[n]p,q[n+1]p,q+7(pq)4[n–1]p,q[n]p,q[n+2]p,q
(pq)4[n–1]p,q[n]p,q[n+2]p,q )x4.
Proof BecauseGpn,q(t–x;x) =Gnp,q(t;x) –x,Gpn,q((t–x)2;x) =Gpn,q(t2;x) – 2xGpn,q(t;x) +x2,
andGpn,q((t–x)4;x) =Gnp,q(t4;x) – 4xGpn,q(t3;x) + 6x2Gpn,q(t2;x) – 4x3Gnp,q(t;x) +x4, and from
Lemma2.2, we obtain Lemma2.3easily.
Lemma 2.4 The sequences(pn), (qn)satisfy0 <qn<pn≤1such that pn→1,qn→1and
pn
n→α,qnn→β, [n]pn,qn→ ∞as n→ ∞,then
lim
n→∞[n– 1]pn,qnGpn
,qn
n (t–x;x) = –
α+β
2 x; (5)
lim
n→∞[n– 1]pn,qnG
pn,qn
n
(t–x)2;x= (α+β)x2; (6)
lim
n→∞[n– 1]pn,qnG
pn,qn
n
Proof Using
lim
n→∞[n– 1]pn,qn
pn qn – 1 – pn qn
pnn+1
[n+ 2]pn,qn
= lim
n→∞[n+ 2]pn,qn
pn qn – 1 – pn qn
pn+1
n
[n+ 2]pn,qn
= lim n→∞
pn+2
n –qnn+2 pn–qn
√
pn–√qn √ qn – pn qn
pnn+1
=α–β
2 –α= –
α+β
2 ,
we get (5) and (6) easily. Letk=n– 2, we have
[n+ 2]pn,qn[n+ 3]pn,qn[n+ 4]pn,qn =q3n[k]pn,qn+p
k n[3]pn,qn
q4n[k]pn,qn+p
k n[4]pn,qn
q5n[k]pn,qn+p
k n[5]pn,qn
∼q12n[k]3pn,qn+pknqn7[5]pn,qn+q8n[4]pn,qn+q9n[3]pn,qn
[k]2pn,qn.
Similarly, we can obtain
[n– 1]pn,qn[n+ 2]pn,qn[n+ 3]pn,qn∼q7n[k]3pn,qn+p
k n
q3
n[4]pn,qn+q4n[3]pn,qn
[k]2
pn,qn, [n– 1]pn,qn[n]pn,qn[n+ 2]pn,qn∼q
4
n[k]3pn,qn+p
k n
q3n+qn[3]pn,qn
[k]2pn,qn, [n– 1]pn,qn[n]pn,qn[n+ 1]pn,qn∼q
3
n[k]3pn,qn+p
k n
q2n+qn[2]pn,qn
[k]2pn,qn.
By Lemma2.3, we can have
Gpn,qn
n
(t–x)4;x∼
An+
1 [k]pn,qn
Bn
x4,
whereAn=q12n – 4p
5 2
nq
19 2
n – 4p
9 2
nq
15 2
n + 7p4nq8nand Bn=pkn
q7n[5]pn,qn+q
8
n[4]pn,qn+q
9
n[3]pn,qn– 4(pnqn)
5 2q3
n[4]pn,qn+q
4
n[3]pn,qn
– 4(pnqn)
9 2q2
n+qn[2]pn,qn
+ 7(pnqn)4
q3n+qn[3]pn,qn
.
SetP=√pn,Q=√qn, by
An=P24– 4P5Q19– 4P9Q15+ 7P8Q16 ∼P9– 4P5Q4– 4P9+ 7P8Q
= 3P5P4–Q4–Q4P5–Q5– 7P8(P–Q)
= (P–Q)
3P5
3
i=0
PiQ3–i–Q4
4
i=0
PiQ4–i– 7P8
we easily obtain
[n– 1]pn,qnAn∼[n]pn,qn(P–Q)
3P5
3
i=0
PiQ3–i–Q4
4
i=0
PiQ4–i– 7P8
∼pnn–qnn pn–qn
pn–qn √
pn+√qn
3P5
3
i=0
PiQ3–i–Q4
4
i=0
PiQ4–i– 7P8
∼a–b
2 (3×4 – 5 – 7) = 0.
Similarly,Bn∼5 + 4 + 3 – 4×(4 + 3) – 4×(1 + 2) + 7×(1 + 3) = 0, we obtain (7).
3 Approximation properties of(p,q)-gamma operators
In this section, we research the approximation properties of (p,q)-gamma operators. The following two theorems show approximation properties about Lipschitz functions.
Theorem 3.1 Let0 <q<p≤1 and F be any bounded subset of the interval[0,∞).If
f ∈CB[0,∞)∩LipM(γ,F),then,for all x∈(0,∞),we have
Gpn,q(f;x) –f(x)≤MB(x)
γ
2 + 2dγ(x;F),
where d(x;F)is the distance between x and F defined by d(x;F) =inf{|x–y|:y∈F}.
Proof LetFbe the closure ofFin [0,∞). Using the properties of infimum, there is at least a pointy0∈Fsuch thatd(x;F) =|x–y0|. By the triangle inequality, we can obtain
Gpn,q(f;x) –f(x)≤Gpn,qf(x) –f(t);x
≤Gpn,qf(x) –f(y0);x
+Gpn,qf(t) –f(y0);x
≤MGpn,q|t–y0|γ;x
+Gpn,q|x–y0|γ;x
≤MGpn,q|x–t|γ;x+ 2dγ(x;F).
Choosingk1=γ2 andk2=
2
2–γ and using the well-known Hölder inequality, we have
Gpn,q(f;x) –f(x)≤MGnp,q|x–t|k1γ;xk11Gp,q
n
1k2;xk12 + 2dγ(x;F)
≤MGpn,q(x–t)2;x
γ
2 + 2dγ (x;F)
=MB(x)
γ
2 + 2dγ (x;F).
This completes the proof.
Theorem 3.2 Let0 <q<p≤1.Then,for all f ∈LipM(γ),we have
Proof Using the monotonicity of the operatorsGnp,q and the Hölder inequality, we can
obtain
Gp,q
n (f;x) –f(x)≤Gnp,qf(t) –f(x);x
≤MGp,q n
|t–x|γ;x
=MGnp,q|t–x|2
γ
2;x≤MGp,q n
(t–x)2;x
γ
2 =MBγ2(x).
The third theorem is a direct local approximation theorem for the operatorsGpn,q(f;x).
Theorem 3.3 Let0 <q<p≤1,f ∈CB[0,∞).Then,for every x∈(0,∞),there exists a positive constant C1such that
Gpn,q(f;x) –f(x)≤C1ω2
f;B(x) +A2(x)+ωf;A(x).
Proof Forx∈(0,∞), we consider new operatorsHnp,q(f;x) defined by Hnp,q(f;x) =Gpn,q(f;x) +f(x) –fA(x) +x.
Using the operator above and Lemma2.3, we have
Hnp,q(t–x;x) =Gnp,q(t–x;x) –A(x) = 0.
Letx,t∈(0,∞) andg∈CB2[0,∞). Using Taylor’s expansion, we can obtain
g(t) =g(x) +g(x)(t–x) +
t
x
g(u)(t–u) du.
Hence,
Hnp,q(g;x) –g(x)=g(x)Hnp,q(t–x);x+Hnp,q
t
x
g(u)(t–u) du;x
≤Hnp,q
t
x
g(u)(t–u) du;x
≤Gpn,q
t
x
g(u)(t–u) du;x
–
A(x)+x x
g(u)A(x) +x–udu
≤Gnp,q
t
x
g(u)(t–u) du;x
+
A(x)+x x
g(u)A(x) +x–udu
≤B(x) +A2(x)g.
Using|Gpn,q(f;x)| ≤ f , we have
Gpn,q(f;x) –f(x)
=Hnp,q(f;x) +fA(x) +x– 2f(x)
≤Hnp,q(f –g;x) – (f –g)(x)+Hnp,q(g;x) –g(x)+fA(x) +x–f(x)
Taking infimum over all g∈C2
B[0,∞) and using (3), we can obtain the desired
asser-tion.
The fourth theorem is a result about the rate of convergence for the operatorsGpn,q(f;x):
Theorem 3.4 Let f ∈Cx2[0,∞), 0 <q<p≤1,and a> 0,we have
Gpn,q(f;x) –f(x)C(0,a]≤4Cf
1 +a2B(a) + 2ωa+1
f;B(a).
Proof For allx∈(0,a] andt>a+ 1, we easily have (t–x)2≥(t–a)2≥1, therefore,
f(t) –f(x)≤f(t)+f(x)≤Cf
2 +x2+t2
=Cf
2 +x2+ (x–t–x)2≤Cf
2 + 3x2+ 2(x–t)2
≤Cf
4 + 3x2(t–x)2≤4Cf
1 +a2(t–x)2,
(8)
and for allx∈(0,a],t∈(0,a+ 1], andδ> 0, we have
f(t) –f(x)≤ωa+1
f,|t–x|≤
1 +|t–x|
δ
ωa+1(f;δ). (9)
From (8) and (9), we get
f(t) –f(x)≤4Cf
1 +a2(t–x)2+
1 +|t–x|
δ
ωa+1(f;δ).
By Schwarz’s inequality and Lemma2.3, we have
Gpn,q(f;x) –f(x)
≤Gpn,qf(t) –f(x);x
≤4Cf
1 +a2Gnp,q(t–x)2;x+Gpn,q
1 +|t–x|
δ
;x
ωa+1(f;δ)
≤4Cf
1 +a2Gnp,q(t–x)2;x+ωa+1(f;δ)
1 +1
δ
Gpn,q
(t–x)2;x
≤4Cf
1 +a2B(x) +ωa+1(f;δ)
1 +1
δ
B(x)
≤4Cf
1 +a2B(a) +ω
a+1(f;δ)
1 +1
δ
B(a)
.
By takingδ=√B(a) and supremum over allx∈(0,a], we accomplish the proof of
Theo-rem3.4.
The following three results are theorems about weighted approximation for the opera-torsGpn,q(f;x).
Theorem 3.5 Let f ∈C0x2[0,∞)and the sequences(pn), (qn)satisfy0 <qn<pn≤1such that pn
such that,for all n>N andν> 0,the inequality
sup x∈(0,∞)
|Gpn,qn
n (f;x) –f(x)|
(1 +x2)32+ν
≤4√2Ω
f; 1 [n– 1]pn,qn
(10)
holds.
Proof Fort> 0,x∈(0,∞) andδ> 0, by the definition and properties ofΩ(f;δ), we get
f(t) –f(x)≤1 +x+|x–t|2Ωf;|t–x|
≤21 +x21 + (t–x)21 +|t–x|
δ
Ω(f;δ).
Usingpn
n→1,qnn→1, [n]pn,qn→ ∞asn→ ∞and Lemma2.4, there exists a positive integerN∈N+such that, for alln>N,
Gpn,qn
n
(t–x)2;x≤ 2(1 +x
2)
[n– 1]pn,qn
, (11)
Gpn,qn
n
(t–x)4;x≤1. (12)
SinceGpn,qn
n is linear and positive, we have
Gpn,qn
n (f;x) –f(x)≤2
1 +x2Ω(f;δ)
1 +Gpn,qn
n
(t–x)2;x
+Gpn,qn
n
1 + (t–x)2|t–x|
δ ;x
. (13)
To estimate the second term of (13), applying the Cauchy–Schwarz inequality and (x+
y)2≤2(x2+y2), we have
Gpn,qn
n
1 + (t–x)2|t–x|
δ ;x
≤√2Gpn,qn
n
1 + (t–x)4;x12
Gpn,qn
n
(t–x)2
δ2 ;x
1 2
.
By (11) and (12),
Gpn,qn
n
1 + (t–x)2|t–x|
δ ;x
≤2
√
2(1 +x2)12 δ[n– 1]pn,qn
.
Takingδ=√ 1
[n–1]pn,qn, we can obtain
Gpn,qn
n (f;x) –f(x)≤4 √
21 +x232Ω
f; 1 [n– 1]pn,qn
.
The proof is completed.
Theorem 3.6 Let the sequences(pn), (qn)satisfy0 <qn<pn≤1such that pn→1,qn→1, and pn
n→α,qnn→β, [n]pn,qn→ ∞as n→ ∞.Then,for f∈C
0
x2[0,∞),we have
lim n→∞G
pn,qn
Proof By the Korovkin theorem in [14], we see that it is sufficient to verify the following three conditions:
lim n→∞G
pn,qn
n
tk;x–xkx2= 0, k= 0, 1, 2. (15)
SinceGpn,qn
n (1;x) = 1,Gpnn,qn(t2;x) =x2, then (15) holds true fork= 0, 2. By Lemma2.2, we
can get
Gpn,qn
n (t;x) –xx2= sup
x∈(0,∞)
1 1 +x2G
pn,qn
n (t;x) –x
= sup x∈(0,∞)
x
1 +x2
√pn–√qn √
qn
–
pn qn
pn+1
n
[n+ 2]pn,qn
≤ sup
x∈(0,∞)
√pn–√qn √
qn
–
pn qn
pn+1
n
[n+ 2]pn,qn
→0, n→ ∞.
Thus the proof is completed.
Theorem 3.7 Let the sequences(pn), (qn)satisfy0 <qn<pn≤1such that pn→1,qn→1,
[n]pn,qn→ ∞as n→ ∞.For every f ∈Cx2[0,∞)andκ> 0,we have
lim n→∞x∈sup(0,∞)
|Gpn,qn
n (f;x) –f(x)|
(1 +x2)1+κ = 0.
Proof Letx0∈(0,∞) be arbitrary but fixed. Then
sup x∈(0,∞)
|Gpn,qn
n (f;x) –f(x)|
(1 +x2)1+κ ≤ sup
x∈(0,x0] |Gpn,qn
n (f;x) –f(x)|
(1 +x2)1+κ + sup
x∈(x0,∞) |Gpn,qn
n (f;x) –f(x)|
(1 +x2)1+κ
≤Gpn,qn
n (f;x) –f(x)C(0,x0]
+Cf sup x∈(x0,∞)
|Gpn,qn
n ((1 +t2);x)|
(1 +x2)1+κ + sup
x∈(x0,∞) |f(x)|
(1 +x2)1+κ. (16)
Since|f(x)| ≤Cf(1 +x2), we havesupx∈(x0,∞)
|f(x)|
(1+x2)1+κ ≤
Cf
(1+x2 0)κ
. Let> 0 be arbitrary. We can choosex0to be so large that
Cf
(1 +x20)κ <. (17)
In view of Lemma2.2, whilex∈(x0,∞), we obtain
Cf lim n→∞
|Gpn,qn
n ((1 +t2);x)|
(1 +x2)1+κ =Cf
(1 +x2)
(1 +x2)1+κ =
Cf
(1 +x2)κ ≤
Cf
(1 +x2 0)κ
Using Theorem3.4, we can see that the first term of inequality (16) implies that
Gpn,qn
n (f;x) –f(x)C(0,x0]<, asn→ ∞. (18)
Combining (16)–(18), we get the desired result. The last result is a Voronovskaja-type asymptotic formula for the operatorsGpn,q(f;x).
Theorem 3.8 Let f ∈CB2[0,∞)and the sequences(pn), (qn)satisfy0 <qn<pn≤1such that pn→1,qn→1and pnn→α,qnn→β, [n]pn,qn→ ∞as n→ ∞,where0≤α,β< 1.
Then,for all x∈(0,∞),
lim
n→∞[n– 1]pn,qn
Gpn,qn
n (f;x) –f(x)
=α+β 2
–xf(x) +x2f(x). (19)
Proof Letx∈(0,∞) be fixed. By Taylor’s expansion formula, we obtain
f(t) =f(x) +f(x)(t–x) +
1 2f
(x) +Θ
pn,qn(t,x)
(t–x)2,
where Θpn,qn(x,t) is bounded and limt→xΘpn,qn(t,x) = 0. By applying the operator
Gpn,qn
n (f;x) to the relation above, we obtain
Gpn,qn
n (f;x) –f(x) =f(x)Gpnn,qn
(t–x);x+1 2f
(x)Gpn,qn
n
(t–x)2;x
+Gpn,qn
n
Θpn,qn(t,x)(t–x)
2;x.
Sincelimt→xΘpn,qn(t,x) = 0, then for all> 0, there exists a positive constantδ> 0 which implies|Θpn,qn(t,x)|<for all fixedx∈(0,∞), wherenis large enough, while|t–x| ≤δ, then|Θpn,qn(t,x)|<
C2
δ2(t–x)2, whereC2is a positive constant. Using Lemma2.4, we obtain
[n– 1]pn,qnG
pn,qn
n
Θ(t,x)(t–x)2;x
≤[n– 1]pn,qnG
pn,qn
n
(t–x)2;x
+C2
δ2[n– 1]pn,qnG
pn,qn
n
(t–x)4;x→0 (n→ ∞).
The proof is completed.
Acknowledgements
The authors are thankful to the editor and anonymous referees for their helpful comments and suggestions.
Funding
This research is supported by the National Natural Science Foundation of China (Grant No. 11626031 and Grant No. 11601266), the Philosophy and Social Sciences General Planning Project of Anhui Province of China (Grant No. AHSKYG2017D153), the Natural Science Foundation of Anhui Province of China (Grant No. 1908085QA29), the Natural Science Foundation of Fujian Province of China (Grant No. 2016J05017), the Natural Science Foundation of Anhui Province of China (Grant No. 1908085QA29), the Project for High-level Talent Innovation and Entrepreneurship of Quanzhou (Grant No. 2018C087R), and the Program for New Century Excellent Talents in Fujian Province University. We also thank Fujian Provincial Key Laboratory of Data-Intensive Computing, Fujian University Laboratory of Intelligent Computing and Information Processing and Fujian Provincial Big Data Research Institute of Intelligent Manufacturing of China.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
The authors contributed equally and significantly in writing this paper. All authors read and approved the final manuscript.
Author details
1School of Mathematics and Computation Science, Anqing Normal University, Anhui, China.2Qingdao Hengxing
University of Science and Technology, Shangdong, China. 3School of Mathematics and Computer Science, Quanzhou Normal University, Quanzhou, China.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Received: 26 December 2018 Accepted: 7 April 2019
References
1. Aral, A., Gupta, V.: (p,q)-type beta function of second kind. Adv. Oper. Theor.1(1), 134–146 (2016) 2. Aral, A., Gupta, V., Agarwal, R.P.: Application ofq-Calculus in Operator Theory. Springer, Berlin (2013)
3. Baskakov, V.A.: Primer posledovatel’nosti lineinyh polozitel’nyh operatorov v prostranstve neprerivnyh funkeil (An example of a sequence of linear positive operators in the space of continuous functions). Dokl. Akad. Nauk SSSR113, 249–251 (1957)
4. Berstein, S.: Demonstration du theorems de Weierstrass, fonde sur le probabilities. Commun. Soc. Math. Kharkow.
13(1–2), (1912–1913)
5. Bleimann, G., Butzer, P.L., Hann, L.: A Bernstein-type operator approximating continuous functions on the semi-axis. Indag. Math.42, 255–262 (1980)
6. Cai, Q.B., Zeng, X.M.: On the convergence of a kind ofq-gamma operators. J. Inequal. Appl.2013, 105 (2013) 7. Cai, Q.B., Zeng, X.M.: On the convergence of a kind of a modifiedq-gamma operators. J. Comput. Anal. Appl.15(5),
826–863 (2013)
8. Cai, Q.B., Zhou, G.R.: On (p,q)-analogue of Kantorovich type Bernstein–Stancu–Schurer operators. Appl. Math. Comput.276, 12–20 (2016)
9. Chen, S.N., Cheng, W.T., Zeng, X.M.: Stancu type generalization of modified Gamma operators based onq-integers. Bull. Korean Math. Soc.54(2), 359–373 (2017)
10. Devore, R.A., Lorentz, G.G.: Constructive Approximation. Springer, Berlin (1993)
11. Dogru, O., Gupta, V.: Monotonicity and the asymptotic estimate of Bleimann Butzer and Hahn operators on q-integers. Georgian Math. J.12, 415–422 (2005)
12. Dogru, O., Gupta, V.: Korovkin-type approximation properties of bivariateq-Meyer–König and Zeller operators. Calcolo42, 51–63 (2006)
13. Finta, Z., Gupta, V.: Approximation properties ofq-Baskakov operators. Cent. Eur. J. Math.8(1), 199–211 (2009) 14. Gadzhiev, A.D.: Theorem of the type of P. P. Korovkin type theorems. Mat. Zametki20(5), 781–786 (1976) 15. Govil, N.K., Gupta, V.: Convergence ofq-Meyer–König–Zeller–Durrmeyer operators. Adv. Stud. Contemp. Math.19,
97–108 (2009)
16. Gupta, V.: (p,q)-Szász–Mirakjan–Baskakov operators. Complex Anal. Oper. Theory12(1), 17–25 (2018) 17. Gupta, V., Agarwal, R.P.: Convergence Estimates in Approximation Theory. Springer, New York (2014)
18. Gupta, V., Aral, A.: Convergence of theqanalogue of Szász-beta operators. Appl. Math. Comput.216, 374–380 (2010) 19. Gupta, V., Aral, A.: Some approximation properties ofq-Baskakov Durrmeyer operators. Appl. Math. Comput.218(3),
783–788 (2011)
20. Gupta, V., Radu, C.: Statistical approximation properties ofq-Baskakov Kantorovich operators. Cent. Eur. J. Math.7(4), 809–818 (2009)
21. Gupta, V., Rassias, T.M., Agrawal, P.N., Acu, A.M.: Recent Advances in Constructive Approximation Theory. Springer, New York (2018)
22. Kac, V., Cheung, P.: Quantum Calculus. Springer, New York (2002)
23. Karsli, H.: Rate of convergence of a new Gamma type operators for the functions with derivatives of bounded variation. Math. Comput. Model.45(5–6), 617–624 (2007)
24. Karsli, H., Agrawal, P.N., Goyal, M.: General Gamma type operators based onq-integers. Appl. Math. Comput.251, 564–575 (2015)
25. Karsli, H., Ali, Ö.M.: Direct local and global approximation results for the operators of gamma type. Hacet. J. Math. Stat.
39(2), 241–253 (2010)
26. Korovkin, P.P.: On convergence of linear operators in the space of continuous functions (Russian). Dokl. Akad. Nauk SSSR90, 961–964 (1953)
27. Lupas, A.: Aq-analogue of the Bernstein operator. In: Seminar on Numerical and Statistical Calculus. University of Cluj-Napoca, vol. 9, pp. 85–92 (1987)
28. Mahmoodov, N.I.: Onq-parametric Szász–Mirakjan operators. Mediterr. J. Math.7(3), 297–311 (2010)
29. Mahmoodov, N.I., Gupta, V., Kaffaoglu, H.: On certainq-Phillips operators. Rocky Mt. J. Math.42(4), 1291–1312 (2012) 30. Malik, N., Gupta, V.: Approximation by (p,q)-Baskakov-Beta operators. Appl. Math. Comput.293, 49–53 (2017) 31. Meyer-König, W., Zeller, K.: Bernsteinsche Potenzreihen. Stud. Math.19, 89–94 (1960)
32. Mirakjan, G.M.: Approximation des fonctions continues au moyen de polynomes de la formee–nxk= 0mnC k,nxk
[Approximation of continuous functions with the aid of polynomials of the forme–nxk= 0mnC
k,nxk] (in French).
33. Mursaleen, M., Khan, F., Khan, A.: On (p,q)-analogue of Bernstein operators. Appl. Math. Comput.266, 874–882 (2015) (Erratum: Appl. Math. Comput. 278, 70–71 (2016))
34. Mursaleen, M., Khan, F., Khan, A.: Some approximation results by (p,q)-analogue of Bernstein–Stancu operators. Appl. Math. Comput.264, 392–402 (2015) (Erratum: Appl. Math. Comput. 269, 744–746 (2016))
35. Mursaleen, M., Nasiruzzaman, M., Nurgali, A.: Some approximation results on Bernstein–Schurer operators defined by (p,q)-integers. J. Inequal. Appl.2015, 249 (2015)
36. Phillips, G.M.: Bernstein polynomials based onq-integers. Ann. Numer. Math.4, 511–518 (1997)
37. Szász, O.: Generalization of S. Bernstein’s polynomial to the infinite interval. J. Res. Natl. Bur. Stand.45, 239–245 (1950) 38. Yuksel, I., Ispir, N.: Weighted approximation by a certain family of summation integral-type operators. Comput. Math.
Appl.52(10–11), 1463–1470 (2006)
39. Zhao, C., Cheng, W.T., Zeng, X.M.: Some approximation properties of a kind ofq-gamma Stancu. J. Inequal. Appl.