Startseite A note on the convergence of Phillips operators by the sequence of functions via q-calculus
Artikel Open Access

A note on the convergence of Phillips operators by the sequence of functions via q-calculus

  • Adem Kiliçman ORCID logo , Mohammad Ayman-Mursaleen ORCID logo und Md. Nasiruzzaman ORCID logo EMAIL logo
Veröffentlicht/Copyright: 27. September 2022
Veröffentlichen auch Sie bei De Gruyter Brill

Abstract

The basic aim of this study is to include nonnegative real parameters to allow for approximation findings of the Stancu variant of Phillips operators. We concentrate on the uniform modulus of smoothness in a simple manner before moving on to the approximation in weighted Korovkin’s space. Our study’s goals and outcomes are to fully develop the uniformly approximated findings of Phillips operators. We determine the order of convergence in terms of Lipschitz maximal function and Peetre’s K-functional. In addition, the Voronovskaja-type theorem is also proved.

MSC 2010: 41A25; 41A36; 33C45

1 Introduction and auxiliary results

The Szász positive linear operators are discussed in this article, and approximation properties are given by incorporating the parametric nonnegative variations. We refer to C [ 0 , ) as the class of all continuous functions defined on the interval [ 0 , ) . Szász [1] introduced the following positive linear operators in 1950 for functions f C [ 0 , ) :

(1.1) S τ ( f ; x ) = e τ x k = 0 ( τ x ) k k ! f k τ .

The operators in (1.1) are more powerful and deal with the generalized approximation properties rather than the earlier Bernstein polynomials obtained by Bernstein in 1912 [2]. Sucu [3] recently introduced the Szász operators (1.1) by applying the exponential generating functions in Dunkl form with a new real parameter ϱ on [ 0 , ) . The following new variant of Szász-Dunkl operators given by Içöz and Çekim [4] for ϱ > 1 2 , namely the q -Szász-Dunkl form, is used to discuss the approximation of Szász operators (1.1):

(1.2) D τ , q ( f ; x ) = 1 e ϱ , q ( [ τ ] q x ) k = 0 ( [ τ ] q x ) k γ ϱ , q ( k ) f 1 q 2 ϱ θ k + k 1 q τ

for ϱ > 1 2 , x 0 , 0 < q < 1 , and f C [ 0 , ) . While for ϱ > 1 2 , Cheikh et al. [5] utilized the q -Hermite polynomials and introduced the definitions of q -Dunkl analogue for the exponential generating function for x [ 0 , ) , and they also investigated the recursion formulae as follows:

(1.3) e ϱ , q ( x ) = τ = 0 x τ γ ϱ , q ( τ ) ; E ϱ , q ( x ) = τ = 0 q τ ( τ 1 ) 2 x τ γ ϱ , q ( τ ) ;

(1.4) γ ϱ , q ( τ + 1 ) = q 2 ϱ θ τ + 1 + τ + 1 1 q 1 γ ϱ , q ( τ ) , τ N ,

(1.5) θ τ = 0 for τ 2 J , 1 for τ 2 J + 1 ,

where N = { 1 , 2 , 3 , 4 , } and J = N { 0 } . Moreover, a special formula of γ ϱ , q ( τ ) is mentioned as follows:

(1.6) γ ϱ , q ( τ ) = ( q 2 ϱ + 1 , q 2 ) [ τ + 1 2 ] ( q 2 , q 2 ) [ τ 2 ] ( 1 q ) τ γ τ , q ( τ ) , τ N ,

and some basic calculations for τ N { 0 } in the Dunkl q -integer form are

γ ϱ , q ( 0 ) = 1 , γ ϱ , q ( 1 ) = q 2 ϱ + 1 1 q 1 , γ ϱ , q ( 2 ) = q 2 ϱ + 1 1 q 1 q 2 1 q 1 , γ ϱ , q ( 3 ) = q 2 ϱ + 1 1 q 1 q 2 1 q 1 q 2 ϱ + 3 1 q 1 , γ ϱ , q ( 4 ) = q 2 ϱ + 1 1 q 1 q 2 1 q 1 q 2 ϱ + 3 1 q 1 q 4 1 q 1 .

The q -integer [ τ ] q and q -factorial [ τ ] q ! , respectively, are defined as follows:

(1.7) [ τ ] q = q τ 1 q 1 if q 1 , τ N 1 if q = 1 0 if τ = 0 [ τ ] q ! = 1 if τ = 0 k = 1 τ [ k ] q if τ N .

In the field of approximation theory, the q -calculus has gained a lot of traction and has a lot of applications. It is capable of dealing with all aspects of science, including mathematical science, physical science, computer science, and other disciplines. Basic hypergeometric functions, number theory, harmonic analysis, orthogonal polynomials, combinatorics, quantum theory, mechanics, theory of relativity, and other mathematical topics are among the many applications and individuals linked with the q -calculus (see [6, 7]). Most recently, the notion of q -calculus has been used in summability [8] and approximation theory [9]. By introducing exponential generating functions in multiple extended formulations and having additional applications, the Szász operators have recently been associated with more improvement than the classical operators, for which we refer the reader to see [10,11,12, 13,14,15, 16,17,18, 19,20,21, 22,23,24] and for some related concepts, see [25,26,27].

In this article, we apply the Dunkl method of generating exponential functions in q -calculus to investigate the approximation capabilities of Phillips operators. The Korovkin-type theorem in weighted spaces is obtained by focusing on functions with characteristics of modulus of continuity. We successfully calculate the order of approximation and investigate some direct theorems by applying the Lipschitz class and Peetre’s K -functional. Next, we also discuss here the Voronovskaja-type approximation theorem.

Let { ξ τ , q ( x ) } τ 1 be a sequence of continuous functions defined on semi-axis R + = [ 0 , ) such that

(1.8) ξ τ , q ( x ) = 1 + 4 [ τ ] q 2 x 2 + 1 2 [ τ ] q ϱ , τ N ,

with the notations

(1.9) χ ϱ = χ if χ 0 , 0 if χ < 0 .

We suppose C ζ [ 0 , ) = { f such that f C [ 0 , ) and f ( t ) = O ( t ζ ) , t } , x [ 0 , ) , ζ > τ , τ N { 0 } , and ϱ 1 2 , we define, for all f C ζ [ 0 , ) ,

(1.10) P τ , q ( f ; x ) = [ τ ] q [ τ ] q + λ [ τ ] q e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 Q τ , q ϱ ( x ) 0 1 q e ϱ , q ( [ τ ] q T ) [ τ ] q k + 2 ϱ θ k T k + 2 ϱ θ k [ k + 2 ϱ θ k ] q ! f ( q k + 2 ϱ θ k T ) d q t ,

where 0 γ λ , T = [ τ ] q t + γ [ τ ] q + λ , and

Q τ , q ϱ ( x ) = ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 .

In the case of γ = λ = 0 , we simply obtain that our operators (1.10) are reduced to the operators studied in [28]. In our modifications of generalized operators, we approach the better observations and more appropriate approximation results rather than [28]. In addition, we can say that these types of approximation results are the extended form of the most recent published article by [7,28,29]. If q = 1 , then [29] is reduced to [28], and if γ = λ = 0 , then our operators are reduced to [28], while the article [29] is the generalized formulation of [7].

To estimate the moments, we have successfully used some basic definitions of q -gamma functions.

Definition 1.1

In q -calculus, the fundamental findings of the gamma function are provided by

(1.11) Γ q ( t ) = 0 1 1 q y t 1 E q ( q y ) d q y , t > 1 ,

(1.12) γ q F ( t ) = 0 F ( 1 q ) y t 1 e q ( y ) d q y , t > 1 ,

where Γ q ( t ) = H ( C , t ) γ q C ( t ) and H ( C ; t ) = 1 1 + C C t 1 + 1 C q t ( 1 + C ) q t 1 . Furthermore, we have the following for every positive integer ρ , H ( C ; ρ ) = q ρ ( ρ 1 ) 2 and Γ q ( ρ ) = q ρ ( ρ 1 ) 2 γ q C ( ρ ) , with the gamma mathematical expression as follows:

(1.13) Γ q ( μ + 1 ) = [ μ ] q Γ q ( μ ) for μ > 0 1 for μ = 0 .

B q ( t + 1 , m ) = H ( C , t ) 0 C y t ( 1 + y ) q t + 1 + m d q y = [ t ] q [ m ] q B q ( t , m + 1 ) , t > 0 , m > 0 ,

with

H ( C , t + 1 ) = q t H ( C , t ) ,

H ( C , t + 1 ) = q t ( t + 1 ) 2 , H ( C , 0 ) = 1 .

For q -integers, the improper integral of function h is defined as follows:

0 F h ( y ) d q y = ( 1 q ) ρ N h q ρ F q ρ F , F R { 0 } .

For more details on quantum calculus and some basic information, we propose to see the article [30].

May [7] discovered an inversion formula for the semigroups of positive linear operators in 1954, and the approximation features of Phillips operators were recently examined using the Dunkl generalization of an exponential form, which was executed in [29]. Introducing the Dunkl generalization of q -exponential form [28], which incorporates the generalized approximation properties of Phillips operators for functions f C [ 0 , ) rather than those given in [7,29], has recently improved the approximation results by Phillips operators. Rather than the published articles [7,28,29,31,32], our generalization of Phillips operators provides more relevant and modified convergence features in quantum calculus. Furthermore, in the case of γ = λ = 0 , we find that our new operators simply reduce to the most recently studied operators in [28].

2 Estimation of the associated moments of operators P τ , q

Lemma 2.1

Let the operators P τ , q ( ; ) be defined by (1.10). Then, for any x [ 0 , ) and 0 < q < 1 , we derive the following identities:

( 1 ) P τ , q ( 1 ; x ) = 1 ; ( 2 ) P τ , q ( t ; x ) = ξ τ , q ( x ) + 1 q [ τ ] q ; ( 3 ) P τ , q ( t 2 ; x ) ( q + 1 ) q 3 [ τ ] q 2 + 1 q 2 [ τ ] q ( 2 q + 1 + q 2 [ 1 + 2 ϱ ] q ) ξ τ , q ( x ) + ( ξ τ , q ( x ) ) 2 ; P τ , q ( t 2 ; x ) ( q + 1 ) q 3 [ τ ] q 2 + 1 q 2 [ τ ] q 2 q + 1 + q 2 ( 1 + ϱ ) [ 1 2 ϱ ] q e ϱ , q ( q [ τ ] q ξ τ , q ( x ) ) e ϱ , q ( [ τ ] q x ) ξ τ , q ( x ) + ( ξ τ , q ( x ) ) 2 ; ( 4 ) P τ , q ( t 3 ; x ) ( q + 1 ) ( q 2 + q + 1 ) q 6 [ τ ] q 3 + 1 q 5 [ τ ] q 2 { ( 3 q 3 + 4 q 2 + 3 q + 1 ) + q 2 ( 3 q 2 + 2 q + 1 ) [ 1 + 2 ϱ ] q + q 5 [ 1 + 2 ϱ ] q 2 } ξ τ , q ( x ) + 1 q 4 [ τ ] q { q ( 3 q 2 + 2 q + 1 ) + 3 q 4 [ 1 + 2 ϱ ] q } ( ξ τ , q ( x ) ) 2 + ( ξ τ , q ( x ) ) 3 ;

( 5 ) P τ , q ( t 4 ; x ) ( q + 1 ) ( q 5 + 2 q 4 3 q 3 + 3 q 2 + 2 q + 1 ) q 10 [ τ ] q 4 + 1 q 9 [ τ ] q 3 { ( 4 q 6 + 9 q 5 + 12 q 4 + 12 q 3 + 8 q 2 + 4 q + 1 ) + q 2 ( 6 q 5 + 9 q 4 + 9 q 3 + 7 q 2 + 3 q + 1 ) [ 1 + 2 ϱ ] q + q 5 ( 4 q 3 + 3 q 2 + 2 q + 1 ) [ 1 + 2 ϱ ] q 2 + q 9 [ 1 + 2 ϱ ] q 3 } ξ τ , q ( x ) + 1 q 8 [ τ ] q 2 { q ( 6 q 5 + 9 q 4 + 9 q 3 + 7 q 2 + 3 q + 1 ) + q 4 ( 4 q 3 + + 3 q 2 + 2 q + 1 ) [ 1 + 2 ϱ ] q + 7 q 8 [ 1 + 2 ϱ ] q 2 } ( ξ τ , q ( x ) ) 2 + 1 q 7 [ τ ] q { q 3 ( 4 q 3 + 3 q 2 + 2 q + 1 ) + 6 q 7 [ 1 + 2 ϱ ] q } ( ξ τ , q ( x ) ) 3 + ( ξ τ , q ( x ) ) 4 .

Proof

By utilizing the above q -gamma function, we have

0 1 q q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 e ϱ , q ( [ τ ] q T ) [ τ ] q k + 2 ϱ θ k T k + 2 ϱ θ k [ k + 2 ϱ θ k ] q ! ( q k + 2 ϱ θ k T ) u d q t = 1 [ τ ] q u + 1 1 [ k + 2 ϱ θ k ] q ! q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 + u ( k + 2 ϱ θ k ) 0 1 q ( [ τ ] q T ) k + 2 ϱ θ k + u e ϱ , q ( [ τ ] q T ) [ τ ] q d q t = 1 [ τ ] q u + 1 1 [ k + 2 ϱ θ k ] q ! q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 + u ( k + 2 ϱ θ k ) 0 1 q T k + 2 ϱ θ k + u e ϱ , q ( T ) d q T = [ τ ] q + λ [ τ ] q 1 [ τ ] q u + 1 1 [ k + 2 ϱ θ k ] q ! q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 + u ( k + 2 ϱ θ k ) 0 1 q T k + 2 ϱ θ k + u e ϱ , q ( T ) d q T = [ τ ] q + λ [ τ ] q 1 [ τ ] q u + 1 1 [ k + 2 ϱ θ k ] q ! q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 + u ( k + 2 ϱ θ k ) γ q 1 ( k + 2 ϱ θ k + u + 1 ) = [ τ ] q + λ [ τ ] q 1 [ τ ] q u + 1 1 [ k + 2 ϱ θ k ] q ! q ( k + 2 ϱ θ k ) ( k + 2 ϱ θ k + 1 ) 2 + u ( k + 2 ϱ θ k ) [ k + 2 ϱ θ k + u ] q ! q ( k + 2 ϱ θ k + u ) ( k + 2 ϱ θ k + u + 1 ) 2 = [ τ ] q + λ [ τ ] q 1 [ τ ] q u + 1 [ k + 2 ϱ θ k + u ] q ! [ k + 2 ϱ θ k ] q ! 1 q u ( u + 1 ) 2 .

For f ( t ) = 1 , we have to put u = 0 , therefore

P τ , q ( 1 ; x ) = [ τ ] q [ τ ] q + λ [ τ ] q e ϱ , q ( [ τ ] q x ) k = 0 ( [ τ ] q x ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k ] q ! [ τ ] q [ k + 2 ϱ θ k ] q ! [ τ ] q + λ [ τ ] q = 1 .

For f ( t ) = t , we have to put u = 1 , therefore

P τ , q ( t ; x ) = [ τ ] q [ τ ] q + λ [ τ ] q e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k + 1 ] q ! q [ τ ] q 2 [ k + 2 ϱ θ k ] q ! [ τ ] q + λ [ τ ] q = 1 q [ τ ] q e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k + 1 ] q = 1 q [ τ ] q e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) + 1 [ τ ] q e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k ] q = ξ τ , q ( x ) + 1 q [ τ ] q .

For f ( t ) = t 2 , we have to put u = 2 , therefore, we have

P τ , q ( t 2 ; x ) = [ τ ] q [ τ ] q + λ [ τ ] q e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k + 2 ] q ! q 3 [ τ ] q 3 [ k + 2 ϱ θ k ] q ! [ τ ] q + λ [ τ ] q = 1 q 3 [ τ ] q 2 e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k + 2 ] q [ k + 2 ϱ θ k + 1 ] q = 1 q 3 [ τ ] q 2 e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) { ( 1 + q ) + q ( 1 + 2 q ) [ k + 2 ϱ θ k ] q + q 3 [ k + 2 ϱ θ k ] q 2 } = ( 1 + q ) q 3 [ τ ] q 2 + ( 1 + 2 q ) q 2 [ τ ] q ξ τ , q ( x ) + 1 [ τ ] q 2 e ϱ , q ( [ τ ] q x ) k = 0 ( [ τ ] q x ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k ] q 2 .

From [4] and by (1.2), we use

[ τ ] q 2 x 2 + q 2 ϱ [ 1 2 ϱ ] q e ϱ , q ( q [ τ ] q ξ τ , q ( x ) ) e ϱ , q ( [ τ ] q x ) [ τ ] q ξ τ , q ( x ) 1 [ τ ] q 2 e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ τ ] q 2 ( ξ τ , q ( x ) ) 2 + [ 1 + 2 ϱ ] q [ τ ] q ξ τ , q ( x ) .

Similarly, for f ( t ) = t 3 and f ( t ) = t 4 , we put u = 3 and u = 4 , therefore

P τ , q ( t 3 ; x ) = 1 q 4 [ τ ] q 3 e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k + 3 ] q [ k + 2 ϱ θ k + 2 ] q [ k + 2 ϱ θ k + 1 ] q

and

P τ , q ( t 4 ; x ) = 1 q 10 [ τ ] q 4 e ϱ , q ( [ τ ] q ξ τ , q ( x ) ) k = 0 ( [ τ ] q ξ τ , q ( x ) ) k γ ϱ , q ( k ) [ k + 2 ϱ θ k + 4 ] q [ k + 2 ϱ θ k + 3 ] q [ k + 2 ϱ θ k + 2 ] q [ k + 2 ϱ θ k + 1 ] q .

A simple calculation leads to the following mathematical expressions:

[ k + 2 ϱ θ k + 3 ] q [ k + 2 ϱ θ k + 2 ] q [ k + 2 ϱ θ k + 1 ] q = ( q + 1 ) ( q 2 + q + 1 ) + { q ( 2 q + 1 ) ( q 2 + q + 1 ) + q 3 ( q + 1 ) } [ k + 2 ϱ θ k ] q + { q 3 ( q 2 + q + 1 ) + q 4 ( 2 q + 1 ) } [ k + 2 ϱ θ k ] q 2 + q 6 [ k + 2 ϱ θ k ] q 3 , [ k + 2 ϱ θ k + 4 ] q [ k + 2 ϱ θ k + 3 ] q [ k + 2 ϱ θ k + 2 ] q [ k + 2 ϱ θ k + 1 ] q = ( q + 1 ) ( q 5 + 2 q 4 + 3 q 3 + 3 q 2 + 2 q + 1 ) + { q ( 2 q + 1 ) ( q 5 + 2 q 4 + 3 q 3 + 3 q 2 + 2 q + 1 ) + q 3 ( q + 1 ) ( 2 q 3 + 2 q 2 + 2 q + 1 ) } [ k + 2 ϱ θ k ] q + { q 3 ( q 5 + 2 q 4 + 3 q 3 + 3 q 2 + 2 q + 1 ) + q 4 ( 2 q + 1 ) ( 2 q 3 + 2 q 2 + 2 q + 1 ) + q 7 ( q + 1 ) } [ k + 2 ϱ θ k ] q 2 + { ( 2 q 3 + 2 q 2 + 2 q + 1 ) q 6 + ( 2 q + 1 ) q 8 } [ k + 2 ϱ θ k ] q 3 + [ k + 2 ϱ θ k ] q 4 q 10 .

By utilizing the results of D τ , q ( f ; x ) obtained by (1.2) for f ( t ) = t 3 and f ( t ) = t 4 (see [4]), we obtain the required result.□

Lemma 2.2

Let χ x = t x . Then, for all x [ 0 , ) R and any 0 < q < 1 , the operators P τ , q ( ; ) have the following properties:

1 P τ , q ( ( χ x ) ; x ) = ξ τ , q ( x ) x + 1 q [ τ ] q . 2 P τ , q ( ( χ x ) 2 ; x ) ( q + 1 ) q 3 [ τ ] q 2 + 1 q 2 [ τ ] q ( 1 + 2 q + q 2 [ 1 + 2 ϱ ] q ) ξ τ , q ( x ) + ( ξ τ , q ( x ) ) 2 + x 2 2 x ξ τ , q ( x ) + 1 q [ τ ] q .

Lemma 2.3

For all x [ 0 , ) τ N and 0 < q < 1 , we have the following inequalities:

1 P τ , q ( ( χ x ) 2 ; x ) x 2 + O 1 [ τ ] q x + O 1 [ τ ] q 2 . 2 P τ , q ( ( χ x ) 4 ; x ) x 4 + O 1 [ τ ] q 3 x 3 + O 1 [ τ ] q 2 x 2 + O 1 [ τ ] q x + O 1 [ τ ] q 4 .

3 Approximation by using Korovkin’s theorem

Korovkin’s theorem [33] has a wide range of applications in mathematical science and other fields of study. Our observation of this section is to provide the approximation properties of new operators P τ , q by (1.10) in order to utilize Korovkin’s theorem. Next, we are also able to give the approximation of these operators in weighted space. For this purpose, we use C β ( R + ) for the class of such functions which are continuous and bounded on semi-axis R + and supremum norm is defined on C β ( R + ) such that f C B ( R + ) = sup x R + f ( x ) . In our investigation, to obtain the convergence theorem, we use the sequence of positive real numbers q = q τ , satisfying 0 < q τ < 1 and

(3.1) lim τ q τ = 1 and lim τ q τ τ = d ,

where the real constant 0 d < 1 . Moreover, we take E = { f : x R + , f ( x ) 1 + x 2 is convergent when x } .

Theorem 3.1

If f C [ 0 , ) E , then for q = q τ , 0 < q τ < 1 satisfying (3.1), we have

lim τ P τ , q τ ( f ; x ) = f ( x )

uniformly in x [ 0 , ) .

Proof

We apply Korovkin’s theorem and verify that lim τ P τ , q τ ( t ν ; x ) = x ν is uniform on [ 0 , ) for ν = 0 , 1 , 2 . If τ , then 1 [ τ ] q 0 . By taking into account Lemma 2.1, it is very easy to conclude that lim τ P τ , q τ ( 1 ; x ) = 1 , lim τ P τ , q τ ( t ; x ) = x , and lim τ P τ , q τ ( t 2 ; x ) = x 2 . This observation completes the proof.□

To give other approximation theorems in the view of Korovkin’s theorem, we recall the weighted spaces in the domain of R + , by the following notations:

P σ ( R + ) = { f such that f ( x ) M f σ ( x ) } , Q σ ( R + ) = { f such that f C [ 0 , ) P σ ( R + ) } , Q σ k ( R + ) = f such that f Q σ ( R + ) and lim x f ( x ) σ ( x ) = α ( α a positive number ) ,

with the weight function σ ( x ) = 1 + x 2 and the positive real numbers M f depending on f , and the norm calculated as f σ = sup x 0 f ( x ) σ ( x ) .

Theorem 3.2

For all f Q σ k ( R + ) and the sequence q = q τ such that 0 < q τ < 1 satisfying (3.1), we obtain

lim τ P τ , q τ ( f ; x ) f σ = 0

uniformly in x [ 0 , ) .

Proof

Take f ( t ) = t τ and f ( t ) Q σ k ( R + ) , use Korovkin’s theorem to easily obtain that P τ , q τ ( t τ ; x ) x τ uniformly, if τ approaches to . Thus, Lemma 2.1 gives P τ , q τ ( 1 ; x ) = 1 , therefore

(3.2) lim τ P τ , q τ ( 1 ; x ) 1 σ = 0 .

And

P τ , q τ ( t ; x ) x σ = sup x [ 0 , ) P τ , q τ ( t ; x ) x 1 + x 2 = sup x [ 0 , ) ξ τ , q τ ( x ) x + 1 q τ [ τ ] q τ 1 + x 2 = sup x [ 0 , ) x 2 + 1 4 [ τ ] q τ 2 x 1 + x 2 + 2 q τ 2 q τ [ τ ] q τ sup x [ 0 , ) 1 1 + x 2 .

Then, clearly, 1 [ τ ] q τ 0 as τ , implying that

(3.3) lim τ P τ , q τ ( t ; x ) x σ = 0 .

In a similar way,

P τ , q τ ( t 2 ; x ) x 2 σ = sup x [ 0 , ) P τ , q τ ( t 2 ; x ) x 2 1 + x 2 = 1 q τ 2 [ τ ] q τ ( 1 + 2 q τ q τ 2 + q τ 2 [ 1 + 2 ϱ ] q τ ) sup x [ 0 , ) x 2 + 1 4 [ τ ] q τ 2 1 + x 2 + 1 2 q τ 3 [ τ ] q τ 2 ( 2 + q τ 2 q τ 2 + q τ 3 q τ 3 [ 1 + 2 ϱ ] q τ ) sup x [ 0 , ) 1 1 + x 2 .

Thus, for 1 [ τ ] q τ 0 , we have

(3.4) lim τ P τ , q τ ( t 2 ; x ) x 2 σ = 0 ,

hence the proof is completed.□

Theorem 3.3

For all φ C m [ 0 , ) , m N , and any μ [ 0 , ) the operators P τ , q satisfy the equality

lim τ sup x [ 0 , ) P τ , q ( φ ; x ) φ ( x ) ( 1 + x 2 ) 1 + μ = 0 ,

where q = q τ be the sequence of positive numbers satisfying 0 < q τ < 1 and C m [ 0 , ) denotes the set of mth order continuous functions.

Proof

Taking into account the inequality φ ( x ) φ σ ( 1 + x 2 ) , then for any positive number x 0 , it is easy to obtain that

lim τ sup x [ 0 , ) P τ , q τ ( g ; x ) g ( x ) ( 1 + x 2 ) 1 + μ sup x x 0 P τ , q τ ( φ ; x ) φ ( x ) ( 1 + x 2 ) 1 + μ + sup x x 0 P τ , q τ ( φ ; x ) φ ( x ) ( 1 + x 2 ) 1 + μ P τ , q τ ( φ ; x ) φ ( x ) σ C [ 0 , x 0 ] + g σ sup x x 0 P τ , q τ ( 1 + t 2 ; x ) φ ( x ) ( 1 + x 2 ) 1 + μ + sup x x 0 φ ( x ) ( 1 + x 2 ) 1 + μ = 1 + 2 + 3 ( suppose ) .

Thus,

(3.5) 3 = sup x x 0 φ ( x ) ( 1 + x 2 ) 1 + μ sup x x 0 φ σ ( 1 + x 2 ) ( 1 + x 2 ) 1 + μ φ σ ( 1 + x 0 2 ) μ .

From Lemma 2.1, we have

lim τ sup x x 0 P τ , q τ ( 1 + t 2 ; x ) 1 + x 2 = 1 .

Now, we suppose that for any given ε > 0 , there exists a new positive integer τ 1 N and τ τ 1 satisfying

sup x x 0 P τ , q τ ( 1 + t 2 ; x ) 1 + x 2 ( 1 + x 0 2 ) μ φ σ ε 3 + 1 ,

and for τ τ 1 ,

(3.6) 2 = g σ sup x x 0 P τ , q τ ( 1 + t 2 ; x ) ( 1 + x 2 ) 1 + μ φ σ ( 1 + x 0 2 ) μ + ε 3 .

Taking into account of (3.5) and (3.6), we see that

2 + 3 2 φ σ ( 1 + x 0 2 ) μ + ε 3 .

Choose very large x 0 such that φ σ ( 1 + x 0 2 ) μ ε 6 , then we obtain

(3.7) 2 + 3 2 ε 3 , for τ τ 1 .

Similarly, for τ τ 2 , we have

(3.8) 1 = P τ , q τ ( φ ; x ) φ ( x ) C [ 0 , x 0 ] ε 3 .

Finally, we take τ 3 = max ( τ 1 , τ 2 ) and by combining (3.7) and (3.8), we obtain

sup x [ 0 , ) P τ , q τ ( φ ; x ) φ ( x ) ( 1 + x 2 ) 1 + μ < ε

which gives the desired proof.□

4 Order of approximation

The present part concentrates on the properties of approximation by means of a one-order modulus of continuity of function ϕ . Let us use the symbol ω χ ( ϕ ; δ ) to represent the one-order modulus of continuity of function ϕ , which gives the maximum oscillation of ϕ in a certain period with a length not exceeding a δ . Let f C B ( R + ) , x [ 0 , ) R , and δ > 0 , then ω χ ( ϕ ; δ ) is given as follows:

(4.1) ω χ ( ϕ ; δ ) = sup 0 < χ x δ ϕ ( t ) ϕ ( x ) ,

such that t , x [ 0 , ) and

(4.2) ϕ ( t ) ϕ ( x ) 1 + χ x δ ω χ ( ϕ ; δ ) .

Theorem 4.1

[34] For the sequences of positive linear operators { P τ } τ 1 : C [ r , s ] C [ u , v ] and [ u , v ] [ r , s ] such that

  1. if ϕ C [ r , s ] and x [ u , v ] , it satisfies that

    P τ ( ϕ ; x ) ϕ ( x ) ϕ ( x ) P τ ( 1 ; x ) 1 + P τ ( 1 ; x ) + 1 δ P τ ( ( χ x ) 2 ; x ) P τ ( 1 ; x ) ω χ ( ϕ ; δ ) ,

  2. if ϕ C [ r , s ] , then, for all x [ u , v ] , one has

    P τ ( ϕ ; x ) ϕ ( x ) f ( x ) P τ ( 1 ; x ) 1 + ϕ ( x ) P τ ( χ x ; x ) + P τ ( ( χ x ) 2 ; x ) P τ ( 1 ; x ) + 1 δ P τ ( ( χ x ) 2 ; x ) ω ( ϕ ; δ ) .

Theorem 4.2

Suppose ϕ C [ 0 , ) , then for every x [ 0 , ) and q = q τ such that 0 < q τ < 1 , operators P τ , q τ satisfy the inequality

P τ , q τ ( ϕ ; x ) ϕ ( x ) 2 ω χ ( ϕ ; δ τ , q τ ( x ) ) ,

where δ = δ τ , q τ ( x ) = P τ , q τ ( ( χ x ) 2 ; x ) .

Proof

Applying Lemma 2.1 and Theorem 4.1, then we obtain that

P τ , q τ ( ϕ ; x ) ϕ ( x ) f ( x ) P τ , q τ ( 1 ; x ) 1 + P τ , q τ ( 1 ; x ) + 1 δ P τ , q τ ( ( χ x ) 2 ; x ) P τ , q τ ( 1 ; x ) ω χ ( ϕ ; δ ) ,

choosing δ = δ τ , q τ ( x ) = P τ , q τ ( ( χ x ) 2 ; x ) then we obtain the result.□

Theorem 4.3

Let the sequence q = q τ satisfy 0 < q τ < 1 . Then, for every ϕ C [ 0 , ) , we obtain

P τ , q τ ( ϕ ; x ) ϕ ( x ) Ω τ , q τ ( x ) ϕ ( x ) + 2 δ τ , q τ ( x ) ω χ ( ϕ ; δ τ , q τ ( x ) ) ,

where δ = δ τ , q τ ( x ) = P τ , q τ ( ( χ x ) 2 ; x ) and

Ω τ , q τ ( x ) = x 2 + 1 4 [ τ ] q τ 2 x + 2 q τ 2 q τ [ τ ] q τ .

Proof

We use Lemmas 2.1 and 2.3 and Theorem 4.1, then we obtain

P τ , q τ ( ϕ ; x ) ϕ ( x ) P τ , q τ ( 1 ; x ) 1 ϕ ( x ) + ϕ ( x ) P τ , q τ ( χ x ; x ) + P τ , q τ ( ( χ x ) 2 ; x ) P τ , q τ ( 1 ; x ) + 1 δ P τ , q τ ( ( χ x ) 2 ; x ) ω χ ( ϕ ; δ ) ,

choosing δ = δ τ , q τ ( x ) = P τ , q τ ( ( χ x ) 2 ; x ) , then we obtain the result.□

In order to prove another theorem in weighted modulus of continuity, we use the results from [35] and for an arbitrary φ C m [ 0 , ) , m N one has

(4.3) ω μ ( φ ; δ ) = sup x [ 0 , ) , μ δ φ ( x + μ ) φ ( x ) ( 1 + μ 2 ) ( 1 + x 2 ) ,

and weighed modulus of continuity satisfying lim δ 0 ω μ ( φ ; δ ) = 0 , and

(4.4) φ ( t ) φ ( x ) 2 1 + χ x δ ( 1 + δ 2 ) ( 1 + x 2 ) ( 1 + ( χ x ) 2 ) ω μ ( φ ; δ ) ,

where x , t [ 0 , ) .

Theorem 4.4

Suppose φ C m [ 0 , ) , x [ 0 , ) , and q = q τ satisfying 0 < q τ < 1 , then

sup x [ 0 , A τ , q τ ( ϱ ) ) P τ , q τ ( φ ; x ) φ ( x ) 1 + x 2 K ( 1 + A τ , q τ ( ϱ ) ) ω μ ( φ ; A τ , q τ ( ϱ ) ) ,

where K = 2 ( 2 + K 1 + K 2 ) > 0 , for K 1 , K 2 > 0 are constants, and A τ , q τ ( ϱ ) depending on parameter ϱ .

Proof

From (4.3) and (4.4) and from Cauchy-Schwarz inequality, we easily obtain that

(4.5) P τ , q τ ( φ ; x ) φ ( x ) 2 ( 1 + δ 2 ) ( 1 + x 2 ) ω μ ( φ ; δ ) 1 + P τ , q τ ( ( χ x ) 2 ; x ) + P τ , q τ ( 1 + ( χ x ) 2 ) χ x δ ; x .

By simplifying, it easily leads to

(4.6) P τ , q τ ( 1 + ( χ x ) 2 ) χ x δ ; x = 1 δ P τ , q τ ( χ x ; x ) + P τ , q τ ( χ x ) 2 χ x δ ; x 1 δ ( P τ , q τ ( χ x ) 2 ; x ) 1 2 + ( P τ , q τ ( ( χ x ) 4 ; x ) ) 1 2 P τ , q τ ( χ x ) 2 δ 2 ; x 1 2 = 1 δ ( P τ , q τ ( χ x ) 2 ; x ) 1 2 { 1 + P τ , q τ ( ( χ x ) 4 ; x ) } .

In view of Lemma 2.3, we can easily conclude that

(4.7) P τ , q τ ( ( χ x ) 2 ; x ) A τ , q τ ( ϱ ) ( x 2 + x + 1 ) K 1 ( x + 1 ) 2 as τ ,

(4.8) P τ , q τ ( ( χ x ) 4 ; x ) B τ , q τ ( ϱ ) ( x 4 + x 3 + x 2 + x + 1 ) K 2 ( x + 1 ) 4 as τ ,

where the constants K 1 > 0 and K 2 > 0 .

A τ , q τ ( ϱ ) = max 2 + 2 q τ 2 q τ 3 [ τ ] q τ 2 , 1 q τ 2 + q τ 2 [ 1 + 2 ϱ ] q τ q τ 2 [ τ ] q τ B τ , q τ ( ϱ ) = max { a τ , q τ , b τ , q τ , c τ , q τ , d τ , q τ } ,

where a τ , q τ = O 1 [ τ ] q τ 4 , b τ , q τ = O 1 [ τ ] q τ 3 , c τ , q τ = O 1 [ τ ] q τ 2 , d τ , q τ = O 1 [ τ ] q τ .

Thus, from inequality (4.5), we obtain

P τ , q τ ( φ ; x ) φ ( x ) 2 ( 1 + δ 2 ) ( 1 + x 2 ) ω μ ( φ ; δ ) 1 + P τ , q τ ( ( χ x ) 2 ; x ) + 1 δ ( P τ , q τ ( χ x ) 2 ; x ) 1 2 { 1 + P τ , q τ ( ( χ x ) 4 ; x ) } 2 ( 1 + δ 2 ) ( 1 + x 2 ) ω μ ( φ ; δ ) 1 + K 1 ( x + 1 ) 2 + 1 δ A τ , q τ ( ϱ ) ( x + 1 ) { 1 + K 2 ( x + 1 ) 2 } .

If we put δ = A τ , q τ ( ϱ ) and take supremum over all x [ 0 , A τ , q τ ( ϱ ) ) , then our result is proved.□

5 Rate of convergence

This section allows us to obtain the result for obtaining the rate of convergence of positive linear operators P τ , q defined by (1.10). From [36], we can easily conclude that for the two positive real parameters γ 1 and γ 2 , one has the well-known Lipschitz-type maximal function

(5.1) Lip η = f C [ 0 , ) : f ( t ) f ( x ) K χ x η ( γ 1 x 2 + γ 2 x + t ) η 2 ; x , t [ 0 , ) ,

where K is a positive constant.

Theorem 5.1

Let q = q τ for 0 < q τ < 1 . Then, for any f Lip η , the operators P τ , q satisfy the inequality

P τ , q τ ( f ; x ) f ( x ) K δ τ , q τ ( x ) γ 1 x 2 + γ 2 x η 2 ,

where δ τ , q τ ( x ) is obtained by Theorem 4.2.

Proof

For all f Lip η , let 0 < η 1 . Suppose η = 1 , then we can see that, if γ 1 , γ 2 0 , ( γ 1 x 2 + γ 2 x + t ) 1 2 ( γ 1 x 2 + γ 2 x ) 1 2 . By applying the Cauchy-Schwarz inequality, we obtain

P τ , q τ ( f ; x ) f ( x ) P τ , q τ ( f ( t ) f ( x ) ; x ) + f ( x ) ( 1 ; x ) 1 P τ , q τ χ x ( γ 1 x 2 + γ 2 x + t ) 1 2 ; x K ( γ 1 x 2 + γ 2 x ) 1 2 P τ , q τ ( χ x ; x ) K ( γ 1 x 2 + γ 2 x ) 1 2 P τ , q τ ( ( χ x ) 2 ; x ) 1 2 .

Thus, the above conclusions hold for η = 1 . Similarly, now we need to verify that the result holds for 0 < η < 1 . For this purpose, we use the monotonicity property and Hölder’s inequality to our operators P τ , q τ . Thus,

P τ , q τ ( f ; x ) f ( x ) P τ , q τ ( f ( t ) f ( x ) ; x ) P τ , q τ f ( t ) f ( x ) 2 η ; x η 2 ( P τ , q τ ( 1 ; x ) ) 2 η 2 K P τ , q τ ( ( χ x ) 2 ; x ) t + γ 1 x 2 + γ 2 x η 2 K ( γ 1 x 2 + γ 2 x ) η 2 { P τ , q τ ( ( χ x ) 2 ; x ) } η 2 P ( γ 1 x 2 + γ 2 z ) η 2 ( P τ , q τ ( χ x ) 2 ; x ) η 2 = K δ τ , q τ ( x ) γ 1 x 2 + γ 2 x η 2 .

This completes the proof.□

Furthermore, we also calculate the inequality in order of another local approximation result for the operators P τ , q in Lipschitz spaces. Let the Lipschitz maximal function f C [ 0 , ) for 0 < η 1 and t , x [ 0 , ) , one has the property from [37]

(5.2) ω η ( f ; x ) = sup t x , t [ 0 , ) f ( t ) f ( x ) χ x η .

Theorem 5.2

Suppose q = q τ for 0 < q τ < 1 , then for all f C [ 0 , ) and x [ 0 , ) , it follows that

P τ , q τ ( f ; x ) f ( x ) ( δ τ , q τ ( x ) ) η 2 ω η ( f ; x ) ,

where ω η ( f ; x ) is obtained by (5.2) and δ τ , q τ ( x ) is defined by Theorem 4.2.

Proof

We apply the Hölder inequality. Thus,

P τ , q τ ( f ; x ) f ( x ) P τ , q τ ( f ( t ) f ( x ) ; x ) ω η ( f ; x ) P τ , q τ ( χ x η ; x ) ω η ( f ; x ) ( P τ , q τ ( 1 ; x ) ) 2 η 2 ( P τ , q τ ( χ x 2 ; x ) ) η 2 = ω η ( f ; x ) ( P τ , q τ ( ( χ x ) 2 ; x ) ) η 2 .

Hence, we obtain our result.□

Let C β [ 0 , ) specify the class of all bounded as well as continuous functions on semi-axis [ 0 , ) satisfying the following:

(5.3) C β 2 [ 0 , ) = { φ C β [ 0 , ) : φ , φ C β [ 0 , ) } ,

(5.4) φ C β 2 [ 0 , ) = φ C β [ 0 , ) + φ C β [ 0 , ) + φ C β [ 0 , ) ,

(5.5) φ C β [ 0 , ) = sup x [ 0 , ) φ ( x ) .

Theorem 5.3

Suppose the operators P τ , q ( ; ) be defined by (1.10). Then, for all φ C β 2 [ 0 , ) , if q = q τ such that q τ ( 0 , 1 ) , we have

P τ , q τ ( φ ; x ) φ ( x ) ( Ω τ , q τ ( x ) + δ τ , q τ ( x ) ) φ C β 2 [ 0 , ) ,

where from Theorem 4.3, Ω τ , q τ ( x ) = P τ , q τ ( ( χ x ) ; x ) = x 2 + 1 4 [ τ ] q τ 2 x + 2 q τ 2 q τ [ τ ] q τ , and δ τ , q τ ( x ) is defined by Theorem 4.2.

Proof

Suppose φ C β 2 [ 0 , ) , then from the Taylor series expansion, we obtain

φ ( t ) = φ ( x ) + φ ( x ) ( χ x ) + φ ( ψ ) ( χ x ) 2 2 , ψ ( x , t ) .

Therefore,

P τ , q τ ( φ ; x ) φ ( x ) = φ ( x ) P τ , q τ ( ( χ x ) ; x ) + φ ( ψ ) 2 P τ , q τ ( ( χ x ) 2 ; x ) ,

thus

P τ , q τ ( φ ; x ) φ ( x ) { Ω τ , q τ ( x ) } φ C β [ 0 , ) + { δ τ , q τ ( x ) } φ C β [ 0 , ) 2

where, from Theorem 4.3, we have

Ω τ , q τ ( x ) = P τ , q τ ( ( χ x ) ; x ) = x 2 + 1 4 [ τ ] q τ 2 x + 2 q τ 2 q τ [ τ ] q τ ,

and in Theorem 4.2, the required value of δ τ , q τ ( x ) is presented.

It is easy to see from (5.4), φ C β [ 0 , ) φ C β 2 [ 0 , ) , and φ C β [ 0 , ) φ C β 2 [ 0 , ) , therefore

P τ , q τ ( φ ; x ) φ ( x ) { Ω τ , q τ ( x ) } φ C β 2 [ 0 , ) + { δ τ , q τ ( x ) } φ C β 2 [ 0 , ) 2 .

This completes the proof.□

6 Direct theorem

In this section, we prove some direct theorems, and for these types of approximation results, we use Peetre’s K -functional introduced by Peetre [38] in 1968 and defined by

(6.1) K P ( f ; δ ) = inf { ( f φ C β [ 0 , ) + δ φ C β 2 [ 0 , ) ) : φ C β 2 [ 0 , ) } ,

and for a positive real number M , one has

K P ( f , δ ) M ω 2 ( f , δ 1 2 ) , δ > 0 ,

(6.2) ω 2 ( f ; δ ) = sup 0 < μ < δ sup x [ 0 , ) f ( x + 2 μ ) 2 f ( x + μ ) + f ( x ) .

Theorem 6.1

Take the sequence of positive numbers q = q τ with 0 < q τ < 1 . Then, for all f C β [ 0 , ) and x [ 0 , ) , it follows that

P τ , q τ ( f ; x ) f ( x ) 2 M ω 2 f ; Ω τ , q τ ( x ) + δ τ , q τ ( x ) 2 + min 1 , Ω τ , q τ ( x ) + δ τ , q τ ( x ) 2 f C β [ 0 , ) ,

where M > 0 is a real constant.

Proof

We use Theorem 5.3. For all φ C β [ 0 , ) , we have

P τ , q τ ( f φ ; x ) f φ . Therefore, it is obvious that

P τ , q τ ( f ; x ) f ( x ) P τ , q τ ( f φ ; x ) + P τ , q τ ( φ ; x ) φ ( x ) + f ( x ) φ ( x ) 2 f φ C β [ 0 , ) + ( Ω τ , q τ ( x ) + δ τ , q τ ( x ) ) φ C β 2 [ 0 , ) = 2 f φ C B [ 0 , ) + Ω τ , q τ ( x ) + δ τ , q τ ( x ) 2 φ C β 2 [ 0 , ) .

By using infimum over all φ C β 2 [ 0 , ) and from equality (6.1), we obtain

P τ , q τ ( f ; x ) f ( x ) 2 K P f ; Ω τ , q τ ( x ) + δ τ , q τ ( x ) 2 .

Now, for an absolute constant M > 0 in [39], we use the relation

K P ( f ; δ ) M { ω 2 ( f ; δ ) + min ( 1 , δ ) f } .

This completes the proof.□

Theorem 6.2

For an arbitrary function φ C β 2 [ 0 , ) , we have

(6.3) Q τ , q ( φ ; x ) = P τ , q ( φ ; x ) + φ ( x ) φ x 2 + 1 4 [ τ ] q 2 + 2 q 2 q [ τ ] q .

If q = q τ , 0 < q τ < 1 , then for all ψ C β 2 [ 0 , ) and from (6.3), we obtain

Q τ , q τ ( ψ ; x ) ψ ( x ) δ τ , q τ ( x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q 2 q τ [ τ ] q τ x 2 ψ ,

where δ τ , q τ ( x ) is defined by Theorem 4.2.

Proof

Let ψ C β 2 [ 0 , ) . Then, it is obvious that Q τ , q τ ( 1 ; x ) = 1 and

Q τ , q τ ( t ; x ) = P τ , q τ ( t ; x ) + x x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ = x .

Since

P τ , q τ ( φ ; x ) φ ,

we have

(6.4) Q τ , q τ ( φ ; x ) P τ , q τ ( φ ; x ) + φ ( x ) + φ x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ 3 φ .

Let ψ C β 2 [ 0 , ) . Then, from the Taylor series, we have

ψ ( t ) = ψ ( x ) + ( χ x ) ψ ( x ) + x t ( t θ ) ψ ( θ ) d θ .

Therefore,

Q τ , q τ ( ψ ; x ) ψ ( x ) = ψ ( x ) Q τ , q τ ( χ x ; x ) + Q τ , q τ x t ( t θ ) ψ ( θ ) d θ ; x = Q τ , q τ x t ( t θ ) ψ ( θ ) d θ ; x = P τ , q τ x t ( t θ ) ψ ( θ ) d θ ; x + x x ( x θ ) ψ ( θ ) d θ ; x x x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ θ ψ ( θ ) d θ ; Q τ , q τ ( ψ ; x ) ψ ( x ) P τ , q τ x t ( t θ ) ψ ( θ ) d θ ; x + x x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ θ ψ ( θ ) d θ .

We know the inequality

x t ( t θ ) ψ ( θ ) d θ ( χ x ) 2 ψ

and

x x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ θ ψ ( θ ) d θ x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x 2 ψ .

Thus, we obtain here that

Q τ , q τ ( ψ ; x ) ψ ( x ) P τ , q τ ( ( χ x ) 2 ; x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x 2 ψ .

This completes the proof.□

Theorem 6.3

Let φ C β 2 [ 0 , ) . Then, for every f C β [ 0 , ) , the operators P τ , q τ defined by (1.10) satisfy

P τ , q τ ( f ; x ) f ( x ) M ω 2 f ; 1 2 Θ τ , q τ ( x ) + min 1 ; 1 4 ( Θ τ , q τ ( x ) ) f C β [ 0 , ) + ω χ ( f ; Δ τ , q τ ( x ) ) ,

where q = q τ with 0 < q τ < 1 and δ τ , q τ ( x ) is defined by Theorem 4.2. Moreover, Θ τ , q τ ( x ) = δ τ , q τ ( x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ 2 and Δ τ , q τ ( x ) = x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x .

Proof

We use Theorem 6.2. Thus, for all φ C β 2 [ 0 , ) and f C β [ 0 , ) , we obtain

P τ , q ( f ; x ) f ( x ) = Q τ , q ( f ; x ) f ( x ) + f x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ f ( x ) Q τ , q ( f φ ; x ) + Q τ , q ( φ ; x ) φ ( x ) + φ ( x ) f ( x ) + f x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ f ( x ) 4 f φ + ω χ f ; x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x + δ τ , q τ ( x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ 2 ψ .

Applying the infimum for all φ C β 2 [ 0 , ) , then from equality (6.1), we easily conclude that

P τ , q ( f ; x ) f ( x ) 4 K P f ; 1 4 δ τ , q τ ( x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ 2 + ω χ f ; x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x M ω 2 f ; 1 2 δ τ , q τ ( x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ 2 + min 1 ; 1 4 δ τ , q τ ( x ) + x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ 2 f C β [ 0 , ) + ω χ f ; x 2 + 1 4 [ τ ] q τ 2 + 2 q τ 2 q τ [ τ ] q τ x .

Thus, we complete the proof here.□

7 Voronovskaja-type approximation theorem

In this section, we establish a quantitative Voronovskaja-type theorem for the operators P τ , q ( f ; x ) .

Theorem 7.1

Let q = q τ and 0 < q τ < 1 . Then, for any f C β [ 0 , ) ,

lim τ P τ , q τ ( φ ; x ) φ ( x ) = 0 ,

where φ ( x ) , φ ( x ) C β [ 0 , ) .

Proof

For φ ( x ) C β [ 0 , ) , using the well-known Taylor’s series expression, we obtain

(7.1) φ ( t ) = φ ( x ) + ( χ x ) φ ( x ) + 1 2 ( χ x ) 2 φ ( x ) + ( χ x ) 2 R x ( t ) ,

where the term R x ( t ) is the remainder term and R x [ 0 , ) , R x ( t ) 0 whenever t x . We apply the positive linear operators P τ , q τ to (7.1), and then using the Cauchy-Schwarz inequality, we have

P τ , q τ ( φ ; x ) φ ( x ) = φ ( x ) P τ , q τ ( χ x ; x ) + φ ( x ) 2 P τ , q τ ( ( χ x ) 2 ; x ) + P τ , q τ ( ( χ x ) 2 R x ( t ) ; x ) φ ( x ) P τ , q τ ( χ x ; x ) + φ ( x ) 2 P τ , q τ ( ( χ x ) 2 ; x ) + P τ , q τ ( ( χ x ) 4 ; x ) P τ , q τ ( R x 2 ( t ) ; x ) .

Since lim τ P τ , q τ ( R x 2 ( t ) ; x ) = 0 , we have

lim τ P τ , q τ ( ( χ x ) 2 R x ( t ) ; x ) = 0 ,

lim τ P τ , q τ ( ( χ x ) ; x ) = 0 ,

and

lim τ P τ , q τ ( ( χ x ) 2 ; x ) = 0 ,

therefore

lim τ [ P τ , q τ ( φ ; x ) φ ( x ) ] = 0 .

Thus, the proof is completed.□

As a consequence of Theorem 7.1, we immediately obtain the corollary.

Corollary 7.2

For any φ C [ 0 , ) , we have

lim τ P τ , q τ ( φ ; x ) φ ( x ) Ω τ , q τ ( x ) φ ( x ) δ τ , q τ ( x ) 2 φ ( x ) = 0 ,

where δ τ , q τ ( x ) is defined by Theorem 4.2 and Ω τ , q τ ( x ) is defined by Theorem 4.3.


;

Acknowledgment

Adem Kilicman and Mohammad Ayman-Mursaleen would like to acknowledge that this research was partially supported by the Ministry of Higher Education under the Fundamental Research Grants Scheme (FRGS) with project number FRGS/2/2014/SG04/UPM/01/1 and having vot number 5524674. The second author is also supported by the Jointly Awarded Research Degree (JADD) program by UPM-UON.

  1. Funding information: Not applicable.

  2. Author contributions: All authors have equal contributions.

  3. Conflict of interest: The authors declare that they have no conflict of interest.

  4. Data availability statement: Data sharing is not applicable to this article as no data sets were generated or analysed during this study.

References

[1] O. Szász, Generalization of S. Bernstein’s polynomials to the infinite interval, J. Res. Natl. Bur. Stand. 45 (1950), 239–245. 10.6028/jres.045.024Suche in Google Scholar

[2] S. N. Bernstein, Démonstration du théoréme de Weierstrass fondée sur le calcul des probabilités, Commun. Kharkov Math. Soc. 13 (1912/13), 1–2. Suche in Google Scholar

[3] S. Sucu, Dunkl analogue of Szász operators, Appl. Math. Comput. 244 (2014), 42–48, https://doi.org/10.1016/j.amc.2014.06.088. Suche in Google Scholar

[4] G. İçöz and B. Çekim, Dunkl generalization of Szász operators via q-calculus, J. Ineq. Appl. 2015 (2015), 284, https://doi.org/10.1186/s13660-015-0809-y. Suche in Google Scholar

[5] B. Cheikh, Y. Gaied, and M. Zaghouani, A q-Dunkl-classical q-Hermite type polynomials, Georgian Math. J. 21 (2014), 125–137, https://doi.org/10.1515/gmj-2014-0022. Suche in Google Scholar

[6] F. H. Jackson, On q-definite integrals, Quart. J. Pure Appl. Math. 15 (1910), 193–203. Suche in Google Scholar

[7] C. P. May, On Phillips operators, J. Approx. Theory 20 (1977), 315–322, https://doi.org/10.1016/0021-9045(77)90078-8. Suche in Google Scholar

[8] M. Mursaleen, S. Tabassum, and R. Fatma, On q-statistical summability method and its properties, Iranian J. Sci. Tech. Trans. A: Sci. 46 (2022), no. 2, 455–460, https://doi.org/10.1007/s40995-022-01285-7. Suche in Google Scholar

[9] M. Ayman-Mursaleen and S. Serra-Capizzano, Statistical convergence via q -calculus and a Korovkin’s type approximation theorem, Axioms 11 (2022), 70, https://doi.org/10.3390/axioms11020070. Suche in Google Scholar

[10] T. Khan, S. A. Khan, and N. Rao, On approximation by bivariate Szász-Gamma type hybrid operators, J. Math. Anal. Appl. 14 (2022), 11–27, https://doi.org/10.54671/bmaa-2022-1-2. Suche in Google Scholar

[11] A. Kiliçman, M. Ayman-Mursaleen, and A. A. H. A. Al-Abied, Stancu type Baskakov-ŤDurrmeyer operators and approximation properties, Mathematics 8 (2020), 1164, https://doi.org/10.3390/math8071164. Suche in Google Scholar

[12] T. Acar, Quantitative q-Voronovskaya and q-Grüss-Voronovskaya-type results for q-Szász operators, Georgian Math. J. 23 (2016), 459–468, https://doi.org/10.1515/gmj-2016-0007. Suche in Google Scholar

[13] A. Alotaibi, M. Nasiruzzaman, and M. Mursaleen, A Dunkl type generalization of Szász operators via post-quantum calculus, J. Inequal. Appl. 2018 (2018), 287, https://doi.org/10.1186/s13660-018-1878-5. Suche in Google Scholar PubMed PubMed Central

[14] R. Aslan, On a Stancu form Szász-Mirakjan-Kantorovich operators based on shape parameter λ, Adv. Studies: Euro-Tbilisi Math. J. 15 (2022), 151–166, https://doi.org/10.3251/asetmj/19322008210. Suche in Google Scholar

[15] M. Ayman-Mursaleen, A. Kilicman, and M. Nasiruzzaman, Approximation by q-Bernstein-Kantorovich operators with shifted knots of real parameters, Filomat 36 (2022), 1179–1194, https://doi.org/10.2298/FIL2204179A. Suche in Google Scholar

[16] Q. B. Cai and R. Aslan, On a new construction of generalized q-Bernstein polynomials based on shape parameter λ, Symmetry 13 (2021), 1919, https://doi.org/10.3390/sym13101919. Suche in Google Scholar

[17] Q. B. Cai, A. Kiliçman, and M. Ayman-Mursaleen, Approximation properties and q-statistical convergence of Stancu type generalized Baskakov-Szász operators, J. Funct. Spaces 2022 (2022), 2286500, https://doi.org/10.1155/2022/2286500. Suche in Google Scholar

[18] G. V. Milovanovic, M. Mursaleen, and M. Nasiruzzaman, Modified Stancu type Dunkl generalization of Szász-Kantorovich operators, Rev. Real Acad. Cienc. Exactas Fis. Nat. Ser. A-Mat. RACSAM 112 (2018), 135–151, https://doi.org/10.1007/s13398-016-0369-0. Suche in Google Scholar

[19] M. Mursaleen, M. Nasiruzzaman, and A. Alotaibi, On modified Dunkl generalization of Szász operators via q-calculus, J. Inequal. Appl. 2017 (2017), 38, https://doi.org/10.1186/s13660-017-1311-5. Suche in Google Scholar PubMed PubMed Central

[20] M. Nasiruzzaman and A. F. Aljohani, Approximation by Szász-Jakimovski-Leviatan type operators via aid of Appell polynomials, J. Funct. Spaces 2020 (2020), 9657489, https://doi.org/10.1155/2020/9657489. Suche in Google Scholar

[21] F. Özger, H. M. Srivastava, and S. A. Mohiuddine, Approximation of functions by a new class of generalized Bernstein-Schurer operators, Rev. Real Acad. Cienc. Exactas Fis. Nat. Ser. A-Mat. RACSAM 114 (2020), 173, https://doi.org/10.1007/s13398-020-00903-6. Suche in Google Scholar

[22] N. Rao, M. Heshamuddin, M. Shadab, and A. Srivastava, Approximation properties of bivariate Szász Durrmeyer operators via Dunkl analogue, Filomat 35 (2021), 4515–4532, https://doi.org/10.2298/FIL2113515R. Suche in Google Scholar

[23] N. Rao, A. Wafi, and A. M. Acu, q-Szász-Durrmeyer type operators based on Dunkl analogue, Complex Anal. Oper. Theory 13 (2019), 915–934, https://doi.org/10.1007/s11785-018-0816-3. Suche in Google Scholar

[24] A. Wafi and N. Rao, Szász-Gamma operators based on Dunkl analogue, Iran. J. Sci. Technol. Trans. 43 (2019), 213–223, https://doi.org/10.1007/s40995-017-0433-4. Suche in Google Scholar

[25] S. A. Mohiuddine and F. Özger, Approximation of functions by Stancu variant of Bernstein-Kantorovich operators based on shape parameter α, Rev. Real Acad. Cienc. Exactas Fis. Nat. Ser. A-Mat. RACSAM 114 (2020), 70, https://doi.org/10.1007/s13398-020-00802-w. Suche in Google Scholar

[26] S. A. Mohiuddine, T. Acar, and A. Alotaibi, Construction of a new family of Bernstein-Kantorovich operators, Math. Methods Appl. Sci. 40 (2017), 7749–7759, https://doi.org/10.1002/mma.4559. Suche in Google Scholar

[27] S. A. Mohiuddine, T. Acar, and M. A. Alghamdi, Genuine modified Bernstein-Durrmeyer operators, J. Inequal. Appl. 2018 (2018), 104, https://doi.org/10.1186/s13660-018-1693-z. Suche in Google Scholar PubMed PubMed Central

[28] M. Nasiruzzaman, A. Mukheimer, and M. Mursaleen, Approximation results on Dunkl generalization of Phillips operators via q-calculus, Adv. Differ. Equ. 2019 (2019), 244, https://doi.org/10.1186/s13662-019-2178-1. Suche in Google Scholar

[29] M. Nasiruzzaman and N. Rao, A generalized Dunkl type modifications of Phillips-operators, J. Inequal. Appl. 2018 (2018), 323, https://doi.org/10.1186/s13660-018-1909-2. Suche in Google Scholar PubMed PubMed Central

[30] A. De Sole and V. G. Kac, On integral representation of q-gamma and q-beta functions, Atti Accad. Naz. Lincei Cl. Sci. Fis. Mat. Natur. Rend. Lincei (9) Mat. Appl. 16 (2005), 11–29. Suche in Google Scholar

[31] A. Alotaibi, M. Nasiruzzaman and M. Mursaleen, Approximation by Phillips operators via q-Dunkl generalization based on a new parameter, J. King Saud Univ.-Sci. 33 (2021), 101413, https://doi.org/10.1016/j.jksus.2021.101413. Suche in Google Scholar

[32] Md. Nasiruzzaman, K. J. Ansari, and M. Mursaleen, On the parametric approximation results of Phillips operators involving the q-Appell polynomials, J. Sci. Tech. Trans. A Sci. 46 (2022), 251–263, https://doi.org/10.1007/s40995-021-01219-9. Suche in Google Scholar

[33] F. Altomare, Korovkin type theorems and approximation by positive linear operators, Surv. Approx. Theory 5 (2010), 92–164. Suche in Google Scholar

[34] O. Shisha and B. Bond, The degree of convergence of sequences of linear positive operators, Proc. Nat. Acad. Sci. USA 60 (1968), 1196–1200, https://doi.org/10.1073/pnas.60.4.1196. Suche in Google Scholar PubMed PubMed Central

[35] Ç. Atakut and N. Ispir, Approximation by modified Szász-Mirakjan operators on weighted spaces, Proc. Indian Acad. Sci. Math. Sci. 112 (2002), 571–578, https://doi.org/10.1007/BF02829690. Suche in Google Scholar

[36] M. A. Ozarslan and H. Aktuğlu, Local approximation for certain King type operators, Filomat 27 (2013), 173–181, https://doi.org/10.2298/FIL1301173O. Suche in Google Scholar

[37] B. Lenze, On Lipschitz type maximal functions and their smoothness spaces, Nederl. Akad. Indag. Math. 50 (1988), 53–63, https://doi.org/10.1016/1385-7258(88)90007-8. Suche in Google Scholar

[38] J. Peetre, A theory of interpolation of normed spaces, Noteas de mathematica 39, Instituto de Mathemática Pura e Applicada, Conselho Nacional de Pesquidas, Rio de Janeiro, 1968. Suche in Google Scholar

[39] A. Ciupa, A class of integral Favard-Szász type operators, Stud. Univ. Babeş-Bolyai Math. 40 (1995), 39–47. Suche in Google Scholar

Received: 2022-04-19
Revised: 2022-06-23
Accepted: 2022-08-03
Published Online: 2022-09-27

© 2022 Adem Kiliçman et al., published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Artikel in diesem Heft

  1. Regular Articles
  2. On some summation formulas
  3. A study of a meromorphic perturbation of the sine family
  4. Asymptotic behavior of even-order noncanonical neutral differential equations
  5. Unconditionally positive NSFD and classical finite difference schemes for biofilm formation on medical implant using Allen-Cahn equation
  6. Starlike and convexity properties of q-Bessel-Struve functions
  7. Mathematical modeling and optimal control of the impact of rumors on the banking crisis
  8. On linear chaos in function spaces
  9. Convergence of generalized sampling series in weighted spaces
  10. Persistence landscapes of affine fractals
  11. Inertial iterative method with self-adaptive step size for finite family of split monotone variational inclusion and fixed point problems in Banach spaces
  12. Various notions of module amenability on weighted semigroup algebras
  13. Regularity and normality in hereditary bi m-spaces
  14. On a first-order differential system with initial and nonlocal boundary conditions
  15. On solving pseudomonotone equilibrium problems via two new extragradient-type methods under convex constraints
  16. Local linear approach: Conditional density estimate for functional and censored data
  17. Some properties of graded generalized 2-absorbing submodules
  18. Eigenvalue inclusion sets for linear response eigenvalue problems
  19. Some integral inequalities for generalized left and right log convex interval-valued functions based upon the pseudo-order relation
  20. More properties of generalized open sets in generalized topological spaces
  21. An extragradient inertial algorithm for solving split fixed-point problems of demicontractive mappings, with equilibrium and variational inequality problems
  22. An accurate and efficient local one-dimensional method for the 3D acoustic wave equation
  23. On a weighted elliptic equation of N-Kirchhoff type with double exponential growth
  24. On split feasibility problem for finite families of equilibrium and fixed point problems in Banach spaces
  25. Entire and meromorphic solutions for systems of the differential difference equations
  26. Multiplication operators on the Banach algebra of bounded Φ-variation functions on compact subsets of ℂ
  27. Mannheim curves and their partner curves in Minkowski 3-space E13
  28. Characterizations of the group invertibility of a matrix revisited
  29. Iterates of q-Bernstein operators on triangular domain with all curved sides
  30. Data analysis-based time series forecast for managing household electricity consumption
  31. A robust study of the transmission dynamics of zoonotic infection through non-integer derivative
  32. A Dai-Liao-type projection method for monotone nonlinear equations and signal processing
  33. Review Article
  34. Remarks on some variants of minimal point theorem and Ekeland variational principle with applications
  35. Special Issue on Recent Methods in Approximation Theory - Part I
  36. Coupled fixed point theorems under new coupled implicit relation in Hilbert spaces
  37. Approximation of integrable functions by general linear matrix operators of their Fourier series
  38. Sharp sufficient condition for the convergence of greedy expansions with errors in coefficient computation
  39. Approximation of conic sections by weighted Lupaş post-quantum Bézier curves
  40. On the generalized growth and approximation of entire solutions of certain elliptic partial differential equation
  41. Existence results for ABC-fractional BVP via new fixed point results of F-Lipschitzian mappings
  42. Linear barycentric rational collocation method for solving biharmonic equation
  43. A note on the convergence of Phillips operators by the sequence of functions via q-calculus
  44. Taylor’s series expansions for real powers of two functions containing squares of inverse cosine function, closed-form formula for specific partial Bell polynomials, and series representations for real powers of Pi
  45. Special Issue on Recent Advances in Fractional Calculus and Nonlinear Fractional Evaluation Equations - Part I
  46. Positive solutions for fractional differential equation at resonance under integral boundary conditions
  47. Source term model for elasticity system with nonlinear dissipative term in a thin domain
  48. A numerical study of anomalous electro-diffusion cells in cable sense with a non-singular kernel
  49. On Opial-type inequality for a generalized fractional integral operator
  50. Special Issue on Advances in Integral Transforms and Analysis of Differential Equations with Applications
  51. Mathematical analysis of a MERS-Cov coronavirus model
  52. Rapid exponential stabilization of nonlinear continuous systems via event-triggered impulsive control
  53. Novel soliton solutions for the fractional three-wave resonant interaction equations
  54. The multistep Laplace optimized decomposition method for solving fractional-order coronavirus disease model (COVID-19) via the Caputo fractional approach
  55. Special Issue on Problems, Methods and Applications of Nonlinear Analysis
  56. Some recent results on singular p-Laplacian equations
  57. Infinitely many solutions for quasilinear Schrödinger equations with sign-changing nonlinearity without the aid of 4-superlinear at infinity
  58. Special Issue on Recent Advances for Computational and Mathematical Methods in Scientific Problems
  59. Existence of solutions for a nonlinear problem at resonance
  60. Asymptotic stability of solutions for a diffusive epidemic model
  61. Special Issue on Computational and Numerical Methods for Special Functions - Part I
  62. Fully degenerate Bernoulli numbers and polynomials
  63. Wigner-Ville distribution and ambiguity function associated with the quaternion offset linear canonical transform
  64. Some identities related to degenerate Stirling numbers of the second kind
  65. Two identities and closed-form formulas for the Bernoulli numbers in terms of central factorial numbers of the second kind
  66. λ-q-Sheffer sequence and its applications
  67. Special Issue on Fixed Point Theory and Applications to Various Differential/Integral Equations - Part I
  68. General decay for a nonlinear pseudo-parabolic equation with viscoelastic term
  69. Generalized common fixed point theorem for generalized hybrid mappings in Hilbert spaces
  70. Computation of solution of integral equations via fixed point results
  71. Characterizations of quasi-metric and G-metric completeness involving w-distances and fixed points
  72. Notes on continuity result for conformable diffusion equation on the sphere: The linear case
Heruntergeladen am 10.9.2025 von https://www.degruyterbrill.com/document/doi/10.1515/dema-2022-0154/html
Button zum nach oben scrollen