Skip to main content

A new construction of Lupaş operators and its approximation properties

Abstract

The aim of this paper is to study a new generalization of Lupaş-type operators whose construction depends on a real-valued function ρ by using two sequences \(u_{m} \) and \(v_{m}\) of functions. We prove that the new operators provide better weighted uniform approximation over \([0,\infty )\). In terms of weighted moduli of smoothness, we obtain degrees of approximation associated with the function ρ. Also, we prove Voronovskaya-type theorem, quantitative estimates for the local approximation.

Introduction

The Weierstrass approximation theorem is the basis of approximation theory introduced by Weierstrass [15], which states that each continuous function defined on \([a,b]\) can be approximated uniformly by some polynomial. In 1912, Bernstein [3] established a constructive proof of the Weierstrass theorem by using Korovkin’s theorem [9].

On the other hand, Cárdenas et al. [4] defined the Bernstein-type operators by \(B_{m}(go\rho ^{-1})o\rho \) and also presented a better degree of approximation depending on ρ. This type of approximation operators generalizes the Korovkin set from \(\{e_{0}, e_{1}, e_{2}\}\) to \(\lbrace e_{0}, \rho , \rho ^{2} \rbrace \). In 2014, Aral et al. [1] also proposed a new modification of Szász–Mirakyan type operators to investigate approximation properties of the announced operators acting on functions defined on unbounded intervals \([0,\infty )\). For various other generalizations of Szász–Mirakyan type operators, one can go through these papers [1214] of Srivastava et al.

Very recently, for \(m\geq 1\), \(z\geq 0\), and suitable functions g defined on \([0,\infty )\), Hatice et al. [8] introduced a new modification of Lupaş operators [11] using a suitable function ρ as follows:

$$ L^{\rho }_{m}(g;z)=2^{-m\rho (z)}\sum _{j=0}^{\infty } \frac{(m\rho (z))_{j}}{2^{j}j!} \bigl(go\rho ^{-1} \bigr) \biggl( \frac{j}{m} \biggr), $$
(1.1)

where ρ satisfies following properties:

\((\rho _{1})\):

ρ is a continuously differentiable function on \([0,\infty )\),

\((\rho _{2})\):

\(\rho (0)=0\) and \(\inf_{z\in [0,\infty )}\rho ^{\prime }(z)\geq 1\), and \((m\rho (z))_{j}\) is the rising factorial defined as follows:

$$\begin{aligned}& \bigl(m\rho (z)\bigr)_{0}=1, \\& \bigl(m\rho (z)\bigr)_{j}=\bigl(m\rho (z)\bigr) \bigl(m\rho (z)+1 \bigr) \bigl(m\rho (z)+2\bigr)\cdots \bigl(m\rho (z)+j-1\bigr), \quad j \geq 0. \end{aligned}$$
If we put \(\rho (z)=z\) in (1.1), then it reduces to the classical Lupaş operators defined in [11].

Very recently, a new construction of Szász–Mirakjan operators was given by Aral et al. [2] by using ρ and two sequences of functions \(\alpha _{m}\), \(\beta _{m}\) defined on a subinterval of \([0,\infty )\):

$$ \tilde{S}^{\rho }_{m}(g;z)=e^{-\alpha _{m}(z)}\sum _{j=0}^{\infty } \frac{(\beta _{m}(z))^{j}}{j!} \bigl(go\rho ^{-1} \bigr) \biggl( \frac{j}{m} \biggr). $$
(1.2)

Inspired by the idea used by Aral et al. in [2], in this paper we define a new construction of Lupaş operator (1.1) which depends on \(\alpha _{m}(z)\) and \(\beta _{m}(z)\), where \(\alpha _{m}(z)\) and \(\beta _{m}(z)\) are sequences of functions defined on \(\tilde{E}\subset [0,\infty )\).

The paper is organized as follows. In Sect. 2, the construction of the announced operator is presented and its moments and central moments are calculated. In Sect. 3, we study convergence properties by using weighted space. In Sect. 4, we obtain the rate of convergence of new constructed operators associated with the weighted modulus of continuity. In Sect. 5, we prove Voronovskaya-type asymptotic formula. Finally, in Sect. 6, we give some approximation results related to \(\mathcal{K}\)-functional, also we define a Lipschitz-type functions.

The construction of Lupaş-type operators

Let g be a continuous functions on \([0,\infty )\) and \(\tilde{E}\subset [0,\infty )\). For given \(m_{0}\in \mathbb{N}\), define \(\mathbb{N}_{1}=\{m\in \mathbb{N}:m\geq m_{0}\}\).

Let \(\alpha _{m},\beta _{m}:\tilde{E}\rightarrow \mathbb{R}\) such that

$$ \beta _{m}(z)-\alpha _{m}(z)\geq 0 \quad \text{for any } z\in \tilde{E} \text{ and } m\in \mathbb{N}_{1}, $$
(2.1)

where \(\alpha _{m}\), \(\beta _{m}\) are positive functions defined on .

Then we consider the new operators in the following form:

$$ L^{\rho }_{m}(g;z)=2^{-\alpha _{m}(z)}\sum _{j=0}^{\infty } \frac{(\beta _{m}(z))_{j}}{2^{j}j!} \bigl(go\rho ^{-1} \bigr) \biggl( \frac{j}{m} \biggr), $$
(2.2)

where ρ is a function which satisfies the conditions \((\rho _{1})\) and \((\rho _{2})\).

We will impose some assumptions on these operators, to show the sequence of operators (2.2) is an approximation process.

We suppose that, for \(z\in \tilde{E}\),

$$ L^{\rho }_{m}(1;z)=1+ u_{m}(z), $$

where \(u_{m}:\tilde{E}\rightarrow \mathbb{R}\). From (2.2), we obtain

$$\begin{aligned} L^{\rho }_{m}(1;z) =&2^{-\alpha _{m}(z)}\sum _{j=0}^{\infty } \frac{(\beta _{n}(z))_{j}}{2^{j}j!} \\ =& 2^{\beta _{m}(z)-\alpha _{m}(z)}. \end{aligned}$$

Thus, we get

$$ 2^{\beta _{m}(z)-\alpha _{m}(z)}=1+ u_{m}(z). $$
(2.3)

Secondly, we assume that

$$ L^{\rho }_{m}(\rho ;z)=\rho (z) + v_{m}(z), $$
(2.4)

where \(v_{m}:\tilde{E}\rightarrow \mathbb{R}\). From (2.2), we infer

$$\begin{aligned} L^{\rho }_{m}(\rho ;z) =&2^{-\alpha _{m}(z)} \sum_{j=0}^{\infty } \frac{(\beta _{m}(z))_{j}}{2^{j}j!} \frac{j}{m} \\ =& \frac{\beta _{m}(z)}{m}2^{\beta _{m}(z)-\alpha _{m}(z)}. \end{aligned}$$
(2.5)

From (2.4) and (2.5), we obtain

$$ \frac{\beta _{m}(z)}{m}2^{\beta _{m}(z)-\alpha _{m}(z)}=\rho (z) + v_{m}(z). $$
(2.6)

Now combining (2.3) and (2.6), we get

$$ \beta _{m}(z)=m\frac{\rho (z) + v_{m}(z)}{1+ u_{m}(z)}, $$
(2.7)

and from relation (2.3), we can write

$$ \beta _{m}(z)-\alpha _{m}(z)=\ln _{2}\bigl(1+ u_{m}(z)\bigr) $$

and

$$ \alpha _{m}(z)=m\frac{\rho (z) + v_{m}(z)}{1+ u_{m}(z)}-\ln _{2} \bigl(1+ u_{m}(z)\bigr), $$
(2.8)

where \(u_{m}(z)>-1\) for any \(z\in \tilde{E}\) and \(m\in \mathbb{N}_{1}\).

Therefore, as a consequence, operators (2.2) become

$$ L^{\rho }_{m}(g;z)=2^{-m\frac{\rho (z) + v_{m}(z)}{1+ u_{m}(z)}}\bigl(1+ u_{m}(z)\bigr) \sum_{j=0}^{\infty } \frac{ (m\frac{\rho (z) + v_{m}(z)}{1+ u_{m}(z)} )_{j}}{2^{j}j!} \bigl(go\rho ^{-1} \bigr) \biggl(\frac{j}{m} \biggr) $$
(2.9)

for \(m\in \mathbb{N}_{1}\) and for any \(z\in \tilde{E}\).

We can recover some linear positive operators which are already in the literature. From operators (2.9) and for the particular choices of the functions \(u_{m}\), \(v_{m}\), and ρ:

  1. (i)

    If we take \(u_{m}(z)=v_{m}(z)=0\), operators (2.9) turn out to be operators (1.1).

  2. (ii)

    If we take \(u_{m}(z)=v_{m}(z)=0\), \(\rho (z)=z\), operators (2.9) turn out to be the classical Lupaş operators given in [11] by

    $$ L_{m}(g;z)=2^{-mz}\sum_{j=0}^{\infty } \frac{(mz)_{j}}{2^{j}j!}g \biggl( \frac{j}{m} \biggr). $$

Now, in order to obtain weighted approximation processes, we assume that the following inequalities hold:

$$ \bigl\vert u_{m}(z) \bigr\vert \leq u_{m} \quad \text{and} \quad \bigl\vert v_{m}(z) \bigr\vert \leq v_{m}, \quad z\in \tilde{E} $$
(2.10)

such that

$$ \lim_{m\rightarrow \infty }u_{m}= \lim _{m \rightarrow \infty }v_{m}=0. $$
(2.11)

From (2.10) and (2.11) it is clear that \((L^{\rho }_{m}(g;z) )_{m\geq m_{0}}\) is an approximation process on \(\tilde{E}\subset [0,\infty )\).

Now, we give some lemmas which are required to prove our main results.

Lemma 2.1

For the operators \(L^{\rho }_{m}(g;z)\) and for all \(z\in \tilde{E}\), we have:

  1. 1.

    \(L^{\rho }_{m}(1;z)=1+ u_{m}(z)\),

  2. 2.

    \(L^{\rho }_{m}(\rho ;z)=\rho (z) + v_{m}(z)\),

  3. 3.

    \(L^{\rho }_{m}(\rho ^{2};z)= \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ u_{m}(z)}+ \frac{2}{m}(\rho (z) + v_{m}(z))\),

  4. 4.

    \(L^{\rho }_{m}(\rho ^{3};z)= \frac{ (\rho (z) + v_{m}(z) )^{3}}{ ( 1+ u_{m}(z) )^{2}}+ \frac{6 (\rho (z) + v_{m}(z) )^{2}}{m (1+ u_{m}(z) )}+ \frac{6}{m^{2}}(\rho (z) + v_{m}(z))\),

  5. 5.

    \(L^{\rho }_{m}(\rho ^{4};z)= \frac{ (\rho (z) + v_{m}(z) )^{4}}{ ( 1+ u_{m}(z) )^{3}}+ \frac{12 (\rho (z) + v_{m}(z) )^{3}}{m ( 1+ u_{m}(z) )^{2}}+ \frac{36 (\rho (z) + v_{m}(z) )^{2}}{m^{2} (1+ u_{m}(z) )}+ \frac{26}{m^{3}}(\rho (z) + v_{m}(z))\).

Lemma 2.2

By using Lemma 2.1and by the linearity of operators \(L^{\rho }_{m}\), we can acquire the central moments as follows:

  1. 1.

    \(L^{\rho }_{m}(\rho (\zeta )-\rho (z);z) = v_{m}(z)-\rho (z)u_{m}(z)\),

  2. 2.

    \(L^{\rho }_{m}{((\rho (\zeta )-\rho (u))^{2};u)}= \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ u_{m}(z)}+ \frac{2 (\rho (z) + v_{m}(z) ) ( 1-m\rho (z) )}{m}+ (1 + u_{m}(z) )\rho ^{2}(z)\),

  3. 3.

    \(L^{\rho }_{m}{((\rho (\zeta )-\rho (u))^{3};u)}= \frac{ (\rho (z) + v_{m}(z) )^{3}}{ ( 1+ u_{m}(z) )^{2}}+ \frac{3 (\rho (z) + v_{m}(z) )^{2} ( 2-n\rho (z) )}{m(1 + u_{m}(z))}+ \frac{ (\rho (z) + v_{m}(z) ) ( 6-6n\rho (z)+3m^{2}\rho ^{2}(z) )}{m^{2}}- (1 + u_{m}(z) )\rho ^{3}(z)\),

  4. 4.

    \(L^{\rho }_{m}{((\rho (\zeta )-\rho (u))^{4};u)}= \frac{ (\rho (z) + v_{m}(z) )^{4}}{ ( 1+ u_{m}(z) )^{3}}+ \frac{ (\rho (z) + v_{m}(z) )^{3} ( 12-4n\rho (z) )}{m(1 + u_{m}(z))^{2}}+ \frac{ (\rho (z) + v_{m}(z) )^{2} ( 36-24n\rho (z)+6m^{2}\rho ^{2}(z) )}{m^{2}(1 + u_{m}(z))}+ \frac{ (\rho (z) + v_{m}(z) ) ( 26-24n\rho (z)+12m^{2}\rho ^{2}(z)-4m^{3}\rho ^{3}(z) )}{m^{3}}+ (1 + u_{m}(z) )\rho ^{4}(z)\).

Remark 2.3

If we put \(u_{m}(z)=v_{m}(z)=0\) in Lemma 2.1 and Lemma 2.2, we have the same result proved in Lemma 2 and Lemma 3 of [8].

Direct result in weighted space

In this section, by using weighted space, we discuss some convergence properties for the operators \(L^{\rho }_{m}\).

Let \(\Phi (z)=1+\rho ^{2}(z)\) be a weight function and \(\mathcal{B}_{\Phi }[0,\infty )\) be the weighted spaces defined as follows:

$$ \mathcal{B}_{\Phi }[0,\infty )=\bigl\{ g:[0,\infty )\rightarrow \mathbb{R} | \bigl\vert g(z) \bigr\vert \leq \mathcal{K}_{f}\Phi (z), z\geq 0 \bigr\} , $$

where \(\mathcal{K}_{g}\) is a constant and \(\mathcal{B}_{\Phi }[0,\infty )\) is a normed linear space equipped with the norm

$$ \Vert g \Vert _{\Phi }=\sup_{z\in [0,\infty )} \frac{ \vert g(z) \vert }{\Phi (z)}. $$

Also, the subspaces \(\mathcal{C}_{\Phi }[0,\infty )\), \(U_{\Phi }[0,\infty )\) and \(U_{\Phi }[0,\infty ) \) of \(\mathcal{B}_{\Phi }[0,\infty )\) are defined as

$$\begin{aligned}& \mathcal{C}_{\Phi }[0,\infty )=\bigl\{ g\in \mathcal{B}_{\Phi }[0, \infty ): g \text{ is continuous on }[0,\infty )\bigr\} , \\& \mathcal{C}^{*}_{\Phi }[0,\infty )= \biggl\lbrace g\in \mathcal{C}_{ \Phi }[0,\infty ): \lim_{z\rightarrow \infty } \frac{g(z)}{\Phi (z)} = \mathcal{K}_{g} = \text{Constant} \biggr\rbrace , \\& U_{\Phi }[0,\infty )=\biggl\{ g\in \mathcal{C}_{\Phi }[0,\infty ): \frac{g(z)}{\Phi (z)} \text{ is uniformly continuous on }[0, \infty )\biggr\} . \end{aligned}$$

It is obvious that \(\mathcal{C}^{*}_{\Phi }[0,\infty )\subset U_{\Phi }[0,\infty )\subset \mathcal{C}_{\Phi }[0,\infty )\subset \mathcal{B}_{\Phi }[0,\infty )\).

In [6], Gadjiev proved the following results for the weighted Korovkin-type theorems.

Lemma 3.1

([6])

For \({m\geq 1}\), \(\mathcal{G}_{m}: \mathcal{B}_{\Phi }[0,\infty )\rightarrow \mathcal{B}_{\Phi }[0,\infty )\) satisfying

$$ \bigl\vert \mathcal{G}_{m}(\Phi ;z) \bigr\vert \leq \mathcal{K}_{m}\Phi (z), \quad z\geq 0, $$

holds, where \(\mathcal{K}_{m}>0\) is a constant depending on m.

Theorem 3.2

([6])

For \({m\geq 1}\), \(\mathcal{G}_{m}: \mathcal{B}_{\Phi }[0,\infty )\rightarrow \mathcal{B}_{\Phi }[0,\infty )\) satisfies

$$ \lim_{m\rightarrow \infty } \bigl\Vert \mathcal{G}_{m}\rho ^{i}- \rho ^{i} \bigr\Vert _{\Phi }=0, \quad i=0,1,2. $$

Then, for any function \(g\in C^{*}_{\Phi }[0,\infty )\), we obtain

$$ \lim_{m\rightarrow \infty } \bigl\Vert \mathcal{G}_{m}(g)-g \bigr\Vert _{\Phi }=0. $$

Therefore, our result follows.

Theorem 3.3

For each \(g\in C^{*}_{\Phi }[0,\infty )\), the following relation

$$ \lim_{m\rightarrow \infty }\sup_{z\in \tilde{E}} \frac{ \vert L^{\rho }_{m}(g;z)-g(z) \vert }{\Phi (z)}=0 $$

holds, provided that conditions (2.10) and (2.11) are fulfilled.

Proof

Let \(g\in C^{*}_{\Phi }[0,\infty )\). Then \(|g(z)|\leq \mathcal{H}_{g}\Phi (z)\), \(z\geq 0\). \(L^{\rho }_{m}\) being linear and positive, it is monotone. Thus

$$ L^{\rho }_{m}(\Phi ;z)=1+ u_{m}(z)+ \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ u_{m}(z)}+ \frac{2}{m}{\bigl(\rho (z) + v_{m}(z)\bigr)} $$

implies that the operator \(L^{\rho }_{m}\) maps the space \(C_{\Phi }[0,\infty )\) into \(B_{\Phi }[0,\infty )\).

By (2.10) and Lemma 2.1, we have

$$ \lim_{m\rightarrow \infty }\sup_{z\in \tilde{E}} \frac{ \vert L^{\rho }_{m}(\rho ^{r})-\rho ^{r} \vert }{\Phi (z)}=0, \quad r=0,1,2. $$
(3.1)

As we know each \(L^{\rho }_{m}(g;z)\) is defined on . Now, by considering the following sequence of operators, we extend it on \([0,\infty )\):

$$ \mathcal{A}_{m} = \textstyle\begin{cases} L^{\rho }_{m}(g;z), & \text{if } z\in \tilde{E}, \\ g(z), & \text{if } z\in [0,\infty ) \setminus \tilde{E}. \end{cases} $$

Obviously,

$$ \bigl\Vert \mathcal{A}_{m}(g)-g \bigr\Vert _{\Phi }=\sup_{z\in \tilde{E}}\frac{L^{\rho }_{m}(g;z)-g(z)}{\Phi (z)}. $$
(3.2)

By applying 3.2 to the operators \(\mathcal{G}_{m}=\mathcal{A}_{m}\) the claim will be proved.

Hence, we have to prove that

$$ \bigl\Vert \mathcal{A}_{m}\bigl(\rho ^{r}\bigr)-\rho ^{r} \bigr\Vert _{\Phi } \rightarrow 0 \quad \text{as }m \rightarrow \infty , r=0,1,2. $$

Since

$$ \bigl\Vert \mathcal{A}_{m}\bigl(\rho ^{r}\bigr)-\rho ^{r} \bigr\Vert _{\Phi }=\sup_{z\in \tilde{E}} \frac{ \vert L^{\rho }_{m}(\rho ^{r})(z)-\rho ^{r}(z) \vert }{\Phi (z)}, $$

by using (3.1), we have

$$ \lim_{m\rightarrow \infty } \bigl\Vert \mathcal{A}_{m}(g)-g \bigr\Vert _{\Phi }=0. $$

By using (3.2), we get the desired result. □

Rate of convergence

In this section, by using weighted modulus of continuity \(\omega _{\rho }(g;\delta )\), we determine the rate of convergence for \(L^{\rho }_{m}\) which was recently considered by Holhoş [7] as follows:

$$ \omega _{\rho }(g;\delta )=\sup_{z,\zeta \in [0,\infty ), \vert \rho (\zeta )-\rho (z) \vert \leq \delta } \frac{ \vert g(\zeta )-g(z) \vert }{\Phi (\zeta )+\Phi (z)}, \quad \delta >0, $$
(4.1)

where \(g\in \mathcal{C}_{\Phi }[0,\infty )\), with the following properties:

  1. (i)

    \(\omega _{\rho }(g;0)=0\),

  2. (ii)

    \(\omega _{\rho }(g;\delta )\geq 0\), \(\delta \geq 0\) for \(g\in \mathcal{C}_{\Phi }[0,\infty )\),

  3. (iii)

    \(\lim_{\delta \rightarrow 0}\omega _{\rho }(g;\delta )=0 \) for each \(g\in U_{\Phi }[0,\infty )\).

Theorem 4.1

([7])

Let us consider a sequence of positive linear operators \(\mathcal{G}_{m}:\mathcal{C}_{\Phi }[0,\infty )\rightarrow \mathcal{B}_{ \Phi }[0,\infty )\) with

$$\begin{aligned}& \bigl\Vert \mathcal{G}_{m}\bigl(\rho ^{0}\bigr)-\rho ^{0} \bigr\Vert _{\Phi ^{0}} = a_{m}, \end{aligned}$$
(4.2)
$$\begin{aligned}& \bigl\Vert \mathcal{G}_{m}(\rho )-\rho \bigr\Vert _{\Phi ^{\frac{1}{2}}} = b_{m}, \end{aligned}$$
(4.3)
$$\begin{aligned}& \bigl\Vert \mathcal{G}_{m}\bigl(\rho ^{2}\bigr)-\rho ^{2} \bigr\Vert _{\Phi } = c_{m}, \end{aligned}$$
(4.4)
$$\begin{aligned}& \bigl\Vert \mathcal{G}_{m}\bigl(\rho ^{3} \bigr)-\rho ^{3} \bigr\Vert _{\Phi ^{ \frac{3}{2}}} = d_{m}, \end{aligned}$$
(4.5)

where the sequences \(a_{m}\), \(b_{m}\), \(c_{m}\), and \(d_{m}\) converge to zero as \(m\rightarrow \infty \). Then

$$ \bigl\Vert \mathcal{G}_{m}(g)-g \bigr\Vert _{\Phi ^{\frac{3}{2}}}\leq (7+4a_{m}+2c_{m}) \omega _{\rho }(g;\delta _{m})+ \Vert g \Vert _{\Phi }a_{m} $$
(4.6)

for all \(g\in \mathcal{C}_{\Phi }[0,\infty )\), where

$$ \delta _{m}=2\sqrt{(a_{m}+2b_{m}+c_{m}) (1+a_{m})}+ a_{m}+3b_{m}+3c_{m}+ d_{m}. $$

Theorem 4.2

Let us have, for each \(g\in \mathcal{C}_{\Phi }[0,\infty )\),

$$\begin{aligned} \bigl\Vert L^{\rho }_{m}(g)-g \bigr\Vert _{\Phi ^{\frac{3}{2}}} \leq & \biggl( 7+4u_{m}+2 \biggl(v^{2}_{m}+2v_{m}+ \frac{2}{m}+ \frac{2v_{m}}{m} \biggr) \biggr)\omega _{\rho }(g; \delta _{m}) \\ &{}+ \Vert g \Vert _{\Phi }u_{m}, \end{aligned}$$

where

$$\begin{aligned} \delta _{m} =&2\sqrt{ \biggl( u_{m}+4v_{m}+v^{2}_{m}+ \frac{2}{m}+ \frac{2v_{m}}{m} \biggr) (1+u_{m})} \\ &{}+ v^{3}_{m}+6v^{2}_{m}+12v_{m}+ \frac{6v^{2}_{m}}{m}+ \frac{18v_{m}}{m}+\frac{6v_{m}}{m^{2}}+\frac{12}{m}+ \frac{6}{m^{2}}. \end{aligned}$$

Proof

If we calculate the sequences \((a_{m})\), \((b_{m})\), \((c_{m})\), and \((d_{m})\), then by using Lemma 2.1, clearly we have

$$\begin{aligned}& \bigl\Vert L^{\rho }_{m}\bigl(\rho ^{0}\bigr)-\rho ^{0} \bigr\Vert _{\Phi ^{0}}= \sup_{z\in \tilde{E}}u_{m}(z) \leq u_{m}=a_{m}, \\& \bigl\Vert L^{\rho }_{m}(\rho )-\rho \bigr\Vert _{\Phi ^{\frac{1}{2}}}= \sup_{z\in \tilde{E}}\frac{v_{m}(z)}{\sqrt{1+\rho ^{2}}} \leq v_{m}=b_{m}, \end{aligned}$$

and

$$ \bigl\Vert L^{\rho }_{m}\bigl(\rho ^{2}\bigr)-\rho ^{2} \bigr\Vert _{\Phi }\leq v^{2}_{m}+2v_{n}+ \frac{2}{m}+\frac{2v_{m}}{m}=c_{m}. $$

Finally,

$$\begin{aligned} \bigl\Vert L^{\rho }_{m}\bigl(\rho ^{3}\bigr)-\rho ^{3} \bigr\Vert _{\Phi ^{ \frac{3}{2}}} \leq & v^{3}_{m}+3v^{2}_{m}+3v_{m}+ \frac{6v^{2}_{m}}{m}+ \frac{12v_{m}}{m}+\frac{6}{m}+\frac{6}{m^{2}}+ \frac{6v_{m}}{m^{2}}=d_{m}. \end{aligned}$$

Thus conditions (4.1)–(4.5) are satisfied. Now, by Theorem 4.1, we obtain the desired result. □

Remark 4.3

From property \((iii)\) of \(\omega _{\rho }(g;\delta )\) and Theorem 4.2, we have

$$ \lim_{m\rightarrow \infty } \bigl\Vert L^{\rho }_{m}(g)-g \bigr\Vert _{\Phi ^{\frac{3}{2}}}=0 \quad \text{for } g\in U_{\Phi }[0, \infty ). $$

Voronovskaya-type theorem

In this section, we establish Voronovskaya-type result for \(L^{\rho }_{m}\).

Theorem 5.1

Let \(g\in \mathcal{C}_{\Phi }[0,\infty )\), \(z\in [0,\infty )\) and suppose that \((go\rho ^{-1} )^{\prime }\) and \((go\rho ^{-1} )^{\prime \prime }\) exist at \(\rho (z)\). If \((go\rho ^{-1} )^{\prime \prime }\) is bounded on \([0,\infty )\) and

$$ \lim_{m\rightarrow \infty }mz_{m}(z)=\ell _{1}, \qquad \lim_{m\rightarrow \infty }mv_{m}(z)=\ell _{2}, $$

then we have

$$\begin{aligned} \lim_{m\rightarrow \infty }m \bigl[L^{\rho }_{m}(g;z)-g(z) \bigr] =&\rho (z)\ell _{1}+\ell _{2}-\rho (z)\ell _{1} \bigl(go \rho ^{-1} \bigr)^{\prime }\rho (z)+\rho (z) \bigl(go\rho ^{-1} \bigr)^{\prime \prime }\rho (z). \end{aligned}$$

Proof

By using the Taylor expansion of \((go\rho ^{-1} )\) at \(\rho (z)\in [0,\infty )\), there exists a point ζ lying between z and z, we have

$$\begin{aligned} g(\zeta ) =& \bigl(go\rho ^{-1} \bigr) \bigl(\rho (\zeta ) \bigr) \\ =& \bigl(go\rho ^{-1} \bigr) \bigl(\rho (z)\bigr)+ \bigl(go\rho ^{-1} \bigr)^{\prime }\bigl(\rho (z)\bigr) \bigl(\rho (\zeta )-\rho (z) \bigr) \\ &{}+\frac{ (go\rho ^{-1} )^{\prime \prime }(\rho (z)) (\rho (\zeta )-\rho (z) )^{2}}{2} + \lambda _{z}(\zeta ) \bigl(\rho ( \zeta )-\rho (z) \bigr)^{2}, \end{aligned}$$
(5.1)

where

$$ \lambda _{z}(\zeta )={ \frac{ (go\rho ^{-1} )^{\prime \prime }(\rho (\zeta ))- (go\rho ^{-1} )^{\prime \prime }(\rho (z))}{2}}. $$
(5.2)

Therefore, the assumption on g and (5.2) ensure that

$$ \bigl\vert \lambda _{z}(\zeta ) \bigr\vert \leq {\mathcal{K}} \quad \text{for all } \zeta \in [0, \infty ) $$

and \(\lim_{\zeta \rightarrow z}\lambda _{z}(\zeta )=0\). By applying operators (2.9) to (5.1), we obtain

$$\begin{aligned} \bigl[L^{\rho }_{m}(g;z)-g(z) \bigr] =& \bigl(go\rho ^{-1} \bigr)^{\prime }\bigl(\rho (z)\bigr)L^{\rho }_{m} \bigl(\bigl(\rho (\zeta )-\rho (z)\bigr);z \bigr) \\ &{}+\frac{ (go\rho ^{-1} )^{\prime \prime }(\rho (z)) L^{\rho }_{m} ((\rho (\zeta )-\rho (y))^{2};z )}{2} \\ &{}+ L^{\rho }_{m} \bigl(\lambda ^{z}( \zeta ) \bigl(\bigl(\rho (\zeta )- \rho (z) \bigr)^{2}; z\bigr) \bigr). \end{aligned}$$
(5.3)

From Lemmas 2.1 and 2.2, we obtain

$$ \lim_{m\rightarrow \infty }mL^{\rho }_{m} \bigl(\bigl(\rho (\zeta )- \rho (z)\bigr);z \bigr)=\ell _{2}-\rho (z) \ell _{1} $$
(5.4)

and

$$\begin{aligned} \lim_{m\rightarrow \infty }mL^{\rho }_{m} \bigl(\bigl(\rho (\zeta )- \rho (z)\bigr)^{2};z \bigr)=2\rho (z). \end{aligned}$$
(5.5)

Since from (5.2), for every \(\epsilon >0\), \(\lim_{\zeta \rightarrow z}\lambda _{z}(\zeta )=0\). Let \(\delta >0\) such that \(|\lambda _{z}(\zeta )|<\epsilon \) for every \(\zeta \geq 0\). From the Cauchy–Schwarz inequality, we get immediately

$$\begin{aligned} \lim_{m\rightarrow \infty }mL^{\rho }_{m} \bigl( \bigl\vert \lambda _{z}( \zeta ) \bigr\vert \bigl(\rho (\zeta )-\rho (z) \bigr)^{2};z \bigr) \leq & \epsilon \lim_{m\rightarrow \infty }mL^{\rho }_{m} \bigl(\bigl( \rho (\zeta )-\rho (z)\bigr)^{2};z \bigr) \\ &{}+\frac{L}{\delta ^{2}}\lim_{m\rightarrow \infty }L^{\rho }_{m} \bigl(\bigl(\rho (\zeta )-\rho (z)\bigr)^{4};z \bigr). \end{aligned}$$

Since

$$ \lim_{m\rightarrow \infty }mL^{\rho }_{m} \bigl(\bigl(\rho (\zeta )- \rho (z)\bigr)^{4};z \bigr)=0, $$
(5.6)

we obtain

$$ \lim_{m\rightarrow \infty }mL^{\rho }_{m} \bigl( \bigl\vert \lambda _{z}( \zeta ) \bigr\vert \bigl(\rho ( \zeta )-\rho (z) \bigr)^{2};z \bigr)=0. $$
(5.7)

Thus, taking into account equations (5.4), (5.5), and (5.7) to equation (5.3), this proves the theorem. □

Local and global approximation

In order to prove local approximation theorems for the operators, let \(\mathcal{C}_{B}[0,\infty )\) be the space of real-valued continuous and bounded functions g with the norm \(\| \cdot \| \) given by

$$ \Vert g \Vert =\sup_{0\leq z< \infty } \bigl\vert g(z) \bigr\vert . $$

We begin by considering the \(\mathcal{K}\)-functional

$$ \mathcal{K}_{2}(g,\delta )=\inf_{r\in W^{2}}\bigl\{ \Vert g-r \Vert +\delta \bigl\Vert g^{{\prime \prime }} \bigr\Vert \bigr\} , $$

where \(\delta >0\) and \(W^{2}=\{s\in \mathcal{C}_{B}[0,\infty ):r^{{\prime }},r^{{\prime \prime }}\in \mathcal{C}_{B}[0,\infty )\}\). Then, in view of the known result [5], there exists an absolute constant \(\mathcal{D}>0\) such that

$$ L(g,\delta )\leq \mathcal{D}\omega _{2}(g,\sqrt{\delta }). $$
(6.1)

For \(g\in \mathcal{C}_{B} [0,\infty )\), the second order modulus of smoothness is defined as

$$ \omega _{2} (g,\sqrt{\delta })=\sup_{0< h\leq \sqrt{\delta }} \sup _{z \in [0,\infty )} \bigl\vert g(z+2h)-2g(z+h)+g(z) \bigr\vert , $$

and the usual modulus of continuity is defined as

$$ \omega (g,\delta )=\sup_{0< h\leq \delta } \sup_{z\in [0,\infty )} \bigl\vert g(z+h)-g(z) \bigr\vert . $$

Theorem 6.1

There exists an absolute constant \(\mathcal{D}>0\) such that

$$ \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \mathcal{D}\mathcal{K} \bigl(g,\delta _{m}(z) \bigr), $$

where \(g\in \mathcal{C}_{B} [0,\infty )\) and

$$ \delta _{m}(z)= \biggl\{ \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ u_{m}(z)}+ \frac{2 (\rho (z) + v_{m}(z) ) ( 1-m\rho (z) )}{m}+ \bigl(1 + u_{m}(z) \bigr)\rho ^{2}(z) \biggr\} . $$

Proof

Let \(r\in W^{2}\) and \(z,\zeta \in [0,\infty )\). By using Taylor’s formula we have

$$ r(\zeta )= r(z)+ \bigl(ro\rho ^{-1} \bigr)^{\prime } \bigl(\rho (z)\bigr) \bigl(\rho ( \zeta )-\rho (z)\bigr)+ \int ^{\rho (\zeta )}_{\rho (z)}\bigl(\rho (\zeta )-v\bigr) \bigl(ro \rho ^{-1} \bigr)^{\prime \prime }(v)\,dv. $$
(6.2)

By using the equality

$$ \bigl(ro\rho ^{-1} \bigr)^{\prime \prime } \bigl(\rho (z) \bigr)= \frac{r^{\prime \prime }(z)}{ (\rho ^{\prime }(z) )^{2}}-r^{ \prime \prime }(z) \frac{\rho ^{\prime \prime }(z)}{ (\rho ^{\prime }(z) )^{3}}, $$
(6.3)

putting \(v= \rho (z)\) in the last term in equality (6.2), we get

$$\begin{aligned} \int ^{\rho (\zeta )}_{\rho (z)}\bigl(\rho (\zeta )-v\bigr) \bigl(ro \rho ^{-1} \bigr)^{\prime \prime }(v)\,dv =& \int ^{\zeta }_{z}\bigl(\rho (\zeta )- \rho (z)\bigr) \biggl[ \frac{r^{\prime \prime }(z)\rho ^{\prime }(z)-r^{\prime }(z)\rho ^{\prime \prime }(v)}{(\rho ^{\prime }(z))^{2}} \biggr]\,dz \\ =& \int ^{\rho (\zeta )}_{\rho (z)}\bigl(\rho (\zeta )-v\bigr) \frac{r^{\prime \prime }(\rho ^{-1}(v))}{(\rho ^{\prime }(\rho ^{-1}(v))^{2}}{dv} \\ &{}- \int ^{\rho (\zeta )}_{\rho (z)}\bigl(\rho (\zeta )-v\bigr) \frac{r^{\prime }(\rho ^{-1}(v))\rho ^{\prime \prime }(\rho ^{-1}(v))}{(\rho ^{\prime }(\rho ^{-1}(v))^{3}}{dv}. \end{aligned}$$
(6.4)

By applying \(\mathcal{S^{*}}^{\mu ,\lambda }_{m,\rho }\) to (6.2) and also by using Lemma 2.1 and (6.4), we deduce

$$\begin{aligned} L^{\rho }_{m}(r;z) =&r(z)+L^{\rho }_{m} \biggl( \int ^{\rho (\zeta )}_{ \rho (z)}\bigl(\rho (\zeta )-v\bigr) \frac{r^{\prime \prime }(\rho ^{-1}(v))}{(\rho ^{\prime }(\rho ^{-1}(v))^{2}}{dv};z \biggr) \\ &{}-L^{\rho }_{m} \biggl( \int ^{\rho (\zeta )}_{\rho (z)}\bigl(\rho (\zeta )-v\bigr) \frac{r^{\prime }(\rho ^{-1}(v))\rho ^{\prime \prime }(\rho ^{-1}(v))}{(\rho ^{\prime }(\rho ^{-1}(v))^{3}}{dv};z \biggr). \end{aligned}$$

By using conditions (\(\rho _{1}\)) and (\(\rho _{2}\)), we get

$$\begin{aligned} & \bigl\vert L^{\rho }_{m}(r;z)-r(z) \bigr\vert \leq \mathcal{M}^{\rho }_{m,{2}}(z) \bigl( \bigl\Vert r^{\prime \prime } \bigr\Vert + \bigl\Vert r^{\prime } \bigr\Vert \bigl\Vert \rho ^{\prime \prime } \bigr\Vert \bigr), \end{aligned}$$

where

$$ \mathcal{M}^{\rho }_{m,{2}}(z)= L^{\rho }_{m}{ \bigl(\bigl(\rho (\zeta )-\rho (z)\bigr)^{2};z\bigr)}. $$

For all \(g\in \mathcal{C}_{B} [0,\infty )\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(r;z) \bigr\vert \leq & \bigl\Vert go\rho ^{-1} \bigr\Vert \bigl(1 + u_{m}(z) \bigr) \\ \leq & \Vert g \Vert L^{\rho }_{m}(1;z)= \Vert g \Vert . \end{aligned}$$
(6.5)

Hence we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & \bigl\vert L^{\rho }_{m}(g-r;z) \bigr\vert + \bigl\vert L^{\rho }_{m}(r;z)-r(z) \bigr\vert + \bigl\vert r(z)-g(z) \bigr\vert \\ \leq & 2 \Vert g-r \Vert \\ &{}+ \biggl\{ \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ u_{m}(z)}+ \frac{2 (\rho (z) + v_{m}(z) ) ( 1-m\rho (z) )}{m} \\ &{}+ \bigl(1 + u_{m}(z) \bigr)\rho ^{2}(z) \biggr\} \bigl( \bigl\Vert r^{\prime \prime } \bigr\Vert + \bigl\Vert r^{\prime } \bigr\Vert \bigl\Vert \rho ^{\prime \prime } \bigr\Vert \bigr). \end{aligned}$$

If we choose \(\mathcal{D}=\operatorname{max} \{2, \|\rho ^{\prime \prime }\|\}\), then

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & \mathcal{D} \biggl(2 \Vert g-r \Vert + \biggl\{ \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ u_{m}(z)} \\ &{}+\frac{2 (\rho (z) + v_{m}(z) ) ( 1-m\rho (z) )}{m}+ \bigl(1 + u_{m}(z) \bigr)\rho ^{2}(z) \biggr\} \bigl\Vert r^{\prime \prime } \bigr\Vert _{W^{2}} \biggr). \end{aligned}$$

Taking infimum over all \(r\in W^{2}\), we obtain

$$ \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \mathcal{D}\mathcal{K} \bigl(g,\delta _{m}(z) \bigr). $$

 □

Let \(0<\alpha \leq 1\), ρ be a function with conditions (\(\rho _{1}\)), (\(\rho _{2}\)) and \(\mathrm{Lip}_{\mathcal{M}}(\rho (y); \alpha )\), \(\mathcal{H}\geq 0\) satisfying

$$ \bigl\vert g(\xi )-g(z) \bigr\vert \leq \mathcal{H} \bigl\vert \rho (\xi )-\rho (z) \bigr\vert ^{\alpha },\quad y,\xi \geq 0. $$

Moreover, \(\mathcal{Y}\subset [0, \infty )\) is a bounded subset and the function \(g\in \mathrm{Lip}_{\mathcal{M}}(\rho (z); \alpha )\), \(0<\alpha \leq 1\) on \(\mathcal{Y}\) if

$$ \bigl\vert g(\zeta )-g(z) \bigr\vert \leq \mathcal{H}_{\alpha ,g} \bigl\vert \rho (\zeta )-\rho (z) \bigr\vert ^{\alpha },\quad z\in \mathcal{Y} \text{ and } \zeta \geq 0, $$

where \(\mathcal{H}_{\alpha , g}\) is a constant.

Theorem 6.2

Let ρ be a function satisfying conditions (\(\rho _{1}\)), (\(\rho _{2}\)). Then, for any \(g\in \mathrm{Lip}_{\mathcal{H}}(\rho (z); \alpha )\), \(0<\alpha \leq 1\) and for every \(z\in (0,\infty )\), \(m\in \mathbb{N}\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \mathcal{H} \bigl(\delta _{m}(z) \bigr)^{\frac{\alpha }{2}}, \end{aligned}$$
(6.6)

where

$$ \delta _{m}(z)= \biggl\{ \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ z_{m}(z)}+ \frac{2 (\rho (z) + v_{m}(z) ) ( 1-m\rho (z) )}{m}+ \bigl(1 + z_{m}(z) \bigr)\rho ^{2}(z) \biggr\} . $$

Proof

Assume that \(\alpha =1\). Then, for \(g\in \mathrm{Lip}_{\mathcal{M}}(\alpha ; 1)\) and \(z\in (0,\infty )\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & L^{\rho }_{m}\bigl( \bigl\vert g(\zeta )-g(z) \bigr\vert ;z \bigr) \\ \leq & \mathcal{H}L^{\rho }_{m}\bigl( \bigl\vert \rho (\zeta )-g(z) \bigr\vert ;z\bigr). \end{aligned}$$

By applying the Cauchy–Schwarz inequality, we get

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & \mathcal{H} \bigl[L^{\rho }_{m}{\bigl(\bigl(\rho ( \zeta )-\rho (z)\bigr)^{2};z\bigr)} \bigr]^{\frac{1}{2}} \\ \leq & \mathcal{H}\sqrt{\delta _{m}(z)}. \end{aligned}$$

Let us assume that \(\alpha \in (0,1)\). Then, for \(g\in \mathrm{Lip}_{\mathcal{H}}(\alpha ; 1)\) and \(z\in (0,\infty )\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & L^{\rho }_{m}\bigl( \bigl\vert g(\zeta )-g(z) \bigr\vert ;z \bigr) \\ \leq & \mathcal{H}L^{\rho }_{m}\bigl( \bigl\vert \rho (\zeta )-g(z) \bigr\vert ^{\alpha };z\bigr). \end{aligned}$$

By taking \(p=\frac{1}{\alpha }\) and \(q=\frac{1}{1-{\alpha }}\), \(g\in \mathrm{Lip}_{\mathcal{H}}(\rho (z); \alpha )\), and applying Hölder’s inequality, we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & \mathcal{H} \bigl[L^{\rho }_{m}{( \bigl\vert \bigl(\rho (\zeta )-\rho (z) \bigr\vert ;z\bigr)} \bigr]^{\alpha }. \end{aligned}$$

Finally, by applying the Cauchy–Schwarz inequality, we get

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \mathcal{H} \bigl(\delta _{m}(z) \bigr)^{\frac{\alpha }{2}}. \end{aligned}$$

 □

Theorem 6.3

Let ρ be a function satisfying conditions (\(\rho _{1}\)), (\(\rho _{2}\)) and \(\mathcal{Y}\) be a bounded subset of \([0,\infty )\). Then, for any \(g\in \mathrm{Lip}_{\mathcal{H}}(\rho (z); \alpha )\), \(0<\alpha \leq 1\), on \(\mathcal{Y}\) \(\alpha \in (0,1]\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \mathcal{H}_{\alpha ,g} \bigl\{ \bigl(\delta _{m}(z) \bigr)^{\frac{\alpha }{2}}+2\bigl[\rho ^{ \prime }(z)\bigr]^{\alpha } d^{\alpha }(z,\mathcal{Y}) \bigr\} ,\quad z\in [0, \infty ), m\in \mathbb{N}, \end{aligned}$$

where \(d(z,\mathcal{Y})= \inf \{\|z-x\|: x \in \mathcal{Y}\}\) and \(\mathcal{H}_{\alpha , g}\) is a constant depending on α and g, and

$$ \delta _{m}(z)= \biggl\{ {\frac{1}{(m+1)^{2}}\rho ^{2}(z)}+{ \frac{2m-1}{(m+1)^{2}}\rho (z)}+{\frac{1}{3(m+1)^{2}}} \biggr\} . $$

Proof

Let \(\overline{\mathcal{Y}}\) be the closure of \(\mathcal{Y}\) in \([0,\infty )\). Then there exists a point \({z_{0}} \in \overline{\mathcal{Y}}\) such that \(d(z,\mathcal{Y})= |z-{z_{0}}|\).

Using the monotonicity of \(L^{\rho }_{m}\) and the hypothesis of g, we obtain

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & L^{\rho }_{m} \bigl( \bigl\vert g(\zeta )-g(z_{0}) \bigr\vert ;z \bigr) + L^{\rho }_{m} \bigl( \bigl\vert g(z)-g(z_{0}) \bigr\vert ;z \bigr) \\ \leq & \mathcal{H}_{\alpha ,g} \bigl\lbrace L^{\rho }_{m} \bigl( \bigl\vert \rho (\zeta )-\rho (z) \bigr\vert ^{\alpha };z \bigr)+ 2 \bigl\vert \rho (z)-\rho (z_{0}) \bigr\vert ^{ \alpha } \bigr\rbrace . \end{aligned}$$

By using Hölder’s inequality for \(p=\frac{2}{\alpha }\) and \(q=\frac{2}{2-{\alpha }}\), as well as the fact \(|\rho (z)-\rho (z_{0})|=\rho ^{\prime }(z)|\rho (z)-\rho (z_{0})|\) in the last inequality, we get

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \mathcal{H}_{\alpha ,g} \bigl\{ \bigl[L^{\rho }_{m}{\bigl( \bigl(\rho (\zeta )-\rho (z)\bigr)^{2};z\bigr)} \bigr]^{ \frac{1}{2}}+2 \bigl[\rho ^{\prime }(z) \bigl\vert \rho (z)-\rho (z_{0}) \bigr\vert \bigr]^{\alpha } \bigr\} . \end{aligned}$$

Hence, by Lemma 2.2 we get the proof. □

Now, we recall the local approximation given in [10] for \(g\in \mathcal{C}_{B} [0,\infty )\) as follows:

$$\begin{aligned} \widetilde{\omega }^{\rho }_{\alpha }(g;z)=\sup _{\zeta \neq z, \zeta \in (0,\infty )}\frac{ \vert g(\zeta )-g(z) \vert }{ \vert \zeta -z \vert ^{\alpha }}, \quad z\in [0,\infty ) \text{ and } {\alpha }\in (0,1]. \end{aligned}$$
(6.7)

Then we get the next result.

Theorem 6.4

Let \(g\in \mathcal{C}_{B}[0,\infty )\) and \({\alpha }\in (0,1]\). Then, for all \(z\in [0,\infty )\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq \widetilde{\omega }^{\rho }_{ \alpha }(g;z) \bigl(\delta _{m}(z) \bigr)^{\frac{\alpha }{2}}, \end{aligned}$$

where

$$ \delta _{m}(z)= \biggl\{ \frac{ (\rho (z) + v_{m}(z) )^{2}}{1+ z_{m}(z)}+ \frac{2 (\rho (z) + v_{m}(z) ) ( 1-m\rho (z) )}{m}+ \bigl(1 + z_{m}(z) \bigr)\rho ^{2}(z) \biggr\} . $$

Proof

We know that

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & L^{\rho }_{m}\bigl( \bigl\vert g(\zeta )-g(z) \bigr\vert ;z \bigr). \end{aligned}$$

From equation (6.7), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & \widetilde{\omega }^{\rho }_{\alpha }(g;z)L^{ \rho }_{m} \bigl( \bigl\vert \rho (\zeta )-\rho (z) \bigr\vert ^{\alpha };z\bigr). \end{aligned}$$

By applying Hölder’s inequality with \(p=\frac{2}{\alpha }\) and \(q=\frac{2}{2-{\alpha }}\), we have

$$\begin{aligned} \bigl\vert L^{\rho }_{m}(g;z)-g(z) \bigr\vert \leq & \widetilde{\omega }^{\rho }_{\alpha }(g;z) \bigl[L^{\rho }_{m,q}{ \bigl(\bigl(\rho (\zeta )-\rho (z)\bigr)^{2};z\bigr)} \bigr]^{ \frac{\alpha }{2}} \\ \leq & \widetilde{\omega }^{\rho }_{\alpha }(g;z) \bigl( \delta _{m}(z) \bigr)^{\frac{\alpha }{2}}, \end{aligned}$$

which proves the desired result. □

Conclusion. Here, a new construction of the generalized Lupaş operators is constructed. We have investigated convergence properties, order of approximation, Voronovskaja-type results, and quantitative estimates for the local approximation. The constructed operators have better flexibility and rate of convergence which depend on the selection of the function ρ and the sequences \(u_{m}\), \(v_{m}\).

Availability of data and materials

The data and material used to support the findings of this study are included within the article.

References

  1. 1.

    Aral, A., Inoan, D., Rasa, I.: On the generalized Szász–Mirakyan operators. Results Math. 65(3–4), 441–452 (2014)

    MathSciNet  Article  Google Scholar 

  2. 2.

    Aral, A., Ulusoy, G., Deniz, E.: A new construction of Szász–Mirakyan operators. Numer. Algorithms 77, 313–326 (2018)

    MathSciNet  Article  Google Scholar 

  3. 3.

    Bernstein, S.N.: Démonstration du théorème de Weierstrass fondée sur le calcul des probabilités. Commun. Soc. Math. Kharkov 2(13), 1–2 (1912)

    MATH  Google Scholar 

  4. 4.

    Cárdenas-Morales, D., Garrancho, P., Rasa, I.: Bernstein-type operators which preserve polynomials. Comput. Math. Appl. 62(1), 158–163 (2011)

    MathSciNet  Article  Google Scholar 

  5. 5.

    DeVore, R.A., Lorentz, G.G.: Constructive Approximation. Grundlehren Math. Wiss. [Fundamental Principles of Mathematical Sciences]. Springer, Berlin (1993)

    Google Scholar 

  6. 6.

    Gadzhiev, A.D.: A problem on the convergence of a sequence of positive linear operators on unbounded sets, and theorems that are analogous to P. P. Korovkin’s theorem. Dokl. Akad. Nauk SSSR 218, 1001–1004 (1974) (Russian)

    MathSciNet  Google Scholar 

  7. 7.

    Holhoş, A.: Quantitative estimates for positive linear operators in weighted spaces. Gen. Math. 16(4), 99–110 (2008)

    MathSciNet  MATH  Google Scholar 

  8. 8.

    İlarslan, H.G.İ. Aral, A., Başcanbaz-Tunca, G.: Generalized Lupaş operators. AIP Conf. Proc., 1926, 020019 (2018)

    Article  Google Scholar 

  9. 9.

    Korovkin, P.P.: Linear Operators and Approximation Theory. Hindustan Publishing Corporation, Delhi (1960)

    Google Scholar 

  10. 10.

    Lenze, B.: On Lipschitz type maximal functions and their smoothness spaces. Indag. Math. 50, 53–63 (1988)

    MathSciNet  Article  Google Scholar 

  11. 11.

    Lupaş, A.: The approximation by some positive linear operators. In: Muller, M.W., et al. (eds.) Proceedings of the International Dortmund Meeting on Approximation Theory, pp. 201–229. Akademie Verlag, Berlin (1995)

    Google Scholar 

  12. 12.

    Srivastava, H.M., Içöz, G., Çekim, B.: Approximation properties of an extended family of the Szász–Mirakjan beta-type operators. Axioms 8, Article ID 111 (2019)

    Article  Google Scholar 

  13. 13.

    Srivastava, H.M., Mursaleen, M., Alotaibi, A.M., Nasiruzzaman, Md., Al-Abied, A.A.H.: Some approximation results involving the q-Szász–Mirakjan–Kantorovich type operators via Dunkl’s generalization. Math. Methods Appl. Sci. 40, 5437–5452 (2017)

    MathSciNet  Article  Google Scholar 

  14. 14.

    Srivastava, H.M., Zeng, X.-M.: Approximation by means of the Szász–Bézier integral operators. Int. J. Pure Appl. Math. 14, 283–294 (2004)

    MathSciNet  MATH  Google Scholar 

  15. 15.

    Weierstrass, K.: Über die analytische Darstellbarkeit sogenannter willkürlicher Functionen einer reellen Veränderlichen. Sitz.ber. Kön. Preuss. Akad. Wis. Berl. 2, 633–639 (1885)

    MATH  Google Scholar 

Download references

Acknowledgements

The first author(Mohd Qasim) is grateful to the Council of Scientific and Industrial Research (CSIR), India, for providing the Senior Research Fellowship with file no. (09/1172(0001)/2017-EMR-I). Also, the authors 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.

Funding

This work is supported by the Natural Science Foundation of Fujian Province of China (Grant No. 2020J01783), 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.

Author information

Affiliations

Authors

Contributions

The authors carried out the whole manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Qing-Bo Cai.

Ethics declarations

Competing interests

The authors declare that they have no competing interests.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Qasim, M., Khan, A., Abbas, Z. et al. A new construction of Lupaş operators and its approximation properties. Adv Differ Equ 2021, 51 (2021). https://doi.org/10.1186/s13662-020-03143-5

Download citation

MSC

  • 41A10
  • 41A25
  • 41A36

Keywords

  • Positive operators
  • Voronovskaya-type theorem
  • Weighted modulus of continuity
  • Local approximation
\