Abstract
In the present article, we construct -Szász-Mirakjan-Kantorovich-Stancu operators with three parameters . First, the moments and central moments are estimated. Then, local approximation properties of these operators are established via -functionals and Steklov mean in means of modulus of continuity. Also, a Voronovskaja-type theorem is presented. Finally, the pointwise estimates, rate of convergence, and weighted approximation of these operators are studied.
1. Introduction
During this decades, the applications of -calculus transpired as a new area in the field of operator approximation theory. Many researchers constructed and discussed many positive linear operators based on -integers, -exponential functions, -Gamma functions [1], -Beta functions, and so on. Since Mursaleen et al. first constructed -Bernstein operators [2] and -Bernstein-Stancu operators [3], several generalizations of well-known positive linear operators based on -calculus have been introduced and studied (see [4–11]). In [12], Acar first proposed -Szász-Mirakjan operators defined on . In [13], Kara et al. constructed a modified -Szász-Mirakjan as follows: where , and . Certain basic notations of -calculus are mentioned below (for details see [14]): For each real number , -analogue of named is defined by
And for each nonnegative integer , the -integer and -factorial are defined by
The -analogue of the exponential function is defined by
Let be an arbitrary function and . The -Jackson integral [15] was defined by
And the -Jackson integral over an interval can be defined by
We easily know that -Jackson integral (6) is not positive unless it is assumed that is a nondecreasing function. To solve this problem, Acar et al. [16] defined the -integral of the arbitrary function on interval as follows:
It is obvious that integral (6) and integral (7) of on are equivalence.
The Kantorovich modification of positive linear operators on is a method to approximate the Riemann integrable functions. The idea behind the Kantorovich modifications mainly depends on replacing the sample value by (see [17, 18]). By definite integral substitution, we have . However, two Kantorovich modifications may be not equivalence or cannot use definite integral substitution in -calculus and -calculus. For the researches about -Szász-Mirakjan-Kantorovich-operators, we can see [19–21]. Meantime, the idea behind the Stancu modifications mainly depends on replacing the sample value by with two parameters (see [22]). For the researches about the Stancu modification of -operators, we can see [23, 24]. All these achievements motivate us to construct the Stancu and Kantorovich generalizations of -Szász-Mirakjan (1) with three parameters as follows:
Definition 1. For , , , and , the -Szász-Mirakjan-Kantorovich-Stancu operators can be defined by
2. Auxiliary Results
In order to obtain the approximation properties of the operators , we need the following lemmas and corollaries.
Lemma 2. For , , , we have .
Proof. Using (7),
Lemma 3. ([13], Lemma 4) For , , and , we have
The following lemma will tell us the relation between the moment of the operators and the moment of the operators :
Lemma 4. For , , , , , we have the following recursive relation:
Proof. By direct computation, we have Hence, the proof of Lemma 4 is completed.
Then, the following lemma can be obtain immediately:
Lemma 5. For, , , , we have
Lemma 6. Under the condition of Lemma 5, we can easily obtain the following formulas for the first and second central moments:
Lemma 7. The sequences , satisfy , such that , , as ; then for any , , , , we have
Proof. By , we have . Thus, we easily obtain (15) and (16). As , we can rewrite Set . Applying Lemma 4 and , , we can also rewrite Combining , we can obtain we obtain the required result.
Lemma 8. Let be the set of real-valued continuous bounded functions defined on endowed with the norm . Under the condition of Lemma 5, for any , we have
Proof. In view of (8) and Lemma 5, the proof of this lemma can be obtained easily.
3. Local Approximation
In this section, we will establish local approximation theorem for the operators. For any , we consider the following -functional: where and . The usual modulus of continuity and the second-order modulus of smoothness of can be defined as
By ([25], p.177, Theorem 2.4), there exists an absolute positive constant such that
In the meantime, for and , the Steklov mean is defined as
Thus, , and we can write
It is obvious that and . If is continuous, then and
Thus, we have . Similarly, .
Theorem 9. Under the condition of Lemma 7, then for all and , we have
Proof. For any , we have . Applying to both ends and using Lemma 5, we can obtain By using the Chauchy-Schwarz inequality and taking , we have Theorem 9 is proved.
Theorem 10. Under the condition of Lemma 7, then for all and , there exists an absolute positive constant such that
Proof. First, we define the following new positive linear operators as follows: It is apparent from Lemma 5, Lemma 6, and Lemma 8 that Now for any given function and , we write Taylor’s expansion formula as follows: By applying operators to both sides of the above equality, we can obtain Using (32), (33), and the following inequality, we can get By using (32) and (34), we have Taking the infimum on the right-hand side over all and using (24), we complete the proof of Theorem 10.
Theorem 11. Under the condition of Lemma 7, then for all and , we have
Proof. Applying to both sides of the equality , using mean value theorem and the Chauchy-Schwarz inequality and taking , we can obtain
Theorem 12. Under the condition of Lemma 7, if , then
Proof. For , using the Steklov mean function , we can write By Lemma 8 and properties of the Steklov mean, we can obtain By Taylor’s expansion formula, we have Hence, Setting , we can get the desired result.
By the classic Korovkin theorem, we easily get the following corollary:
Corollary 13. Under the condition of Lemma 7, then for all and any , the the sequence converges to uniformly on .
4. Voronovskaja-Type Theorem for
In this section, we show a Voronovskaja-type asymptotic formula for the operators by means of the first, second and fourth central moments.
Theorem 14. Under the condition of Lemma 7, then for all satisfying that exists at a point , we can obtain
Proof. By Taylor’s expansion formula for , we have where Applying L’Hospital’s Rule, Thus, . Consequently, we can write By Schwarz’s inequality, we have We observe that and . Then, it follows in Corollary 13 that Hence, from (17), we can obtain Combining, we complete the proof of Theorem 14.
Corollary 15. Under the condition of Lemma 7, then for all , we have uniformly with respect to any finite interval .
5. Pointwise Estimates
In this section, we establish two pointwise estimates of the operators . First, we compute the rate of convergence locally by using functions belonging to the Lipschitz class. We denote that is in , , if it satisfies the following condition: where is a positive constant depending only on and .
Theorem 16. The sequences , satisfy , and be any bounded subset on . If , then for any , we have where denotes the distance between and .
Proof. Let be the closure of . Using the properties of infimum, and there is at least a point such that . By the triangle inequality By the monotonicity of , we get Applying the well-known Hölder inequality with , , we obtain Second, we will give a local direct estimation of the operators by using the Lipschitz-type maximal function of the order introduced by Lenze [26] as
Theorem 17. The sequences, satisfyand. If, then for any, we have
Proof. Using the equality (61), we obtain By the well-known Hölder inequality, we have Thus, the proof of Theorem 17 is completed.
6. Rate of Convergence
Let be the set of all functions defined on satisfying the condition with an absolute constant which may depend only on . denotes the subspace of all continuous functions with the norm . By , and we denote the subspace of all functions for which is finite. Meantime, we denote the modulus of continuity of on the interval , by
Theorem 18. Let , , and . Then, for all , we have
Proof. For any and , we easily have ; thus and for any , and , we have For (67) and (68), we can get Applying the Cauchy-Schwarz inequality and choosing , we have This completes the proof of Theorem 18.
7. Weighted Approximation
As is known, if is not uniform, the limit may be not true. In [27], Ispir defined the following weighted modulus of continuity: and proved the properties of monotone increasing about as , and the inequality
Theorem 19. Under the condition of Lemma 7, , then for sufficiently large , the inequality holds, where and is a positive constant depending only on and .
Proof. Applying (71) and (72), we can obtain Thus, for any and , the above inequality can be rewritten Applying (16) and (17), there exists sufficiently large such that By Schwarz’s inequality, we can obtain Using as linear and positive and choosing , we can obtain for sufficiently large and , where .
Theorem 20. Under the condition of Lemma 7, then for any , we have
Proof. Applying the Korovkin theorem [28], we only see that it is sufficient to prove the following three conditions: Since , the condition holds for . By Lemma 6, we can obtain Hence, (80) holds for . Similarly, by Lemma 5, we can write for , Thus, (80) holds for . Hence, the proof is completed.
Theorem 21. Under the condition of Lemma 7, then for any and , we have
Proof. Let be arbitrary but fixed. Applying , we have Let . By Lemma 5, there exists , such that for all : Hence Thus Next, for sufficiently large such that . Then, . Applying Corollary 13, there exists , such that for all , Let . Combining (86), (88), and (89), we have Hence, the proof of Theorem 21 is completed.
Data Availability
No data were used to support this study.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work is supported by the National Natural Science Foundation of China (Grant No. 11626031), the Key Natural Science Research Project in Universities of Anhui Province (Grant No. KJ2019A0572 and KJ2020A0503), the Philosophy and Social Sciences General Planning Project of Anhui Province of China (Grant No. AHSKYG2017D153), and the Natural Science Foundation of Anhui Province of China (Grant No. 1908085QA29).