Next Article in Journal
Photonic Topological Insulator Based on Frustrated Total Internal Reflection in Array of Coupled Prism Resonators
Previous Article in Journal
Modular Version of Edge Irregularity Strength for Fan and Wheel Graphs
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Approximation by Operators for the Sheffer–Appell Polynomials

by
Mdi Begum Jeelani
* and
Abeer S. Alnahdi
Department of Mathematics and Statistics, College of Science, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11564, Saudi Arabia
*
Author to whom correspondence should be addressed.
Symmetry 2022, 14(12), 2672; https://doi.org/10.3390/sym14122672
Submission received: 29 September 2022 / Revised: 25 November 2022 / Accepted: 2 December 2022 / Published: 17 December 2022
(This article belongs to the Section Mathematics)

Abstract

:
In this paper, we introduce a generalization of the Kantrovich–Stancu-type Szasz operator asymmetry with hybrid families of special polynomials. Additionally, we construct certain positive linear operators together with the Sheffer–Appell polynomial sequences and then obtain the properties of convergence and the order of convergence, which is symmetric to these operators. For applications, we consider certain explicit examples including mixed-type special polynomials.

1. Introduction and Preliminaries

In recent years, approximation theory has contributed to the development of different computational techniques as it provides a crucial link between pure and applied mathematics. It deals with the process of approximating the functions in the best way with much simpler or amenable functions and methods depending on using recent approximation processes. In this theory, positive approximation techniques play a fundamental role and emerge in a very natural way in many problems relating to the approximation of the continuous functions, especially when one needs further qualitative properties, such as monotonicity, convexity, shape preservation, symmetry, and so on.
The positive estimate processes given by Korovkin [1] play out as an essential technique in order to determine several related practical and symphonic investigations, measure hypotheses, PDEs, and probability hypotheses. In 1953, P. P. Korovkin [1] discovered perhaps the most powerful and, at the same time, the simplest criterion in order to decide whether a given sequence ( K n ) n N of positive linear operators on the space C [ 0 , 1 ] is an approximation process, i.e., K n ( f ) f uniformly on [ 0 , 1 ] for every f C [ 0 , 1 ] . Following this outcome, many mathematicians have extended Korovkin’s theorem to other function spaces or, more generally, to abstract spaces, such as Banach lattices, Banach algebras, Banach spaces, and so on. Korovkin’s work, in fact, delineated a new theory that may be called the Korovkin-type approximation theory. The Korovokin-type approximation properties and convergence rates have been inspected by numerous scientists, e.g., see [1,2,3,4,5,6]. Some interesting contributions to approximation theories can be found in the following studies [7,8].
The Szasz operators [5] are popular examples of positive linear operators.
Szasz [5] proposed the following positive linear operators:
S η ( f ˇ ; t ) : = e η t κ = 0 ( η t ) κ κ ! f κ η ,
where t 0 and f ˇ C [ 0 , ) once the sum (1) converges. Many researchers have used the generalization of Szasz operators, see for example [3,6,8,9].
In 1880, Appell [10] established and studied η t h -degree sequences of polynomials A η ( t ) , η = 0 , 1 , 2 , . These polynomials satisfy the recurrence relation
d d t A η ( t ) = η A η 1 ( t ) , η = 0 , 1 , 2 , ,
and have the generating function as follows:
A ( x ) e t x = κ = 0 A κ ( t ) x κ ,
where
A ( x ) = κ = 0 θ κ x κ , θ 0 0 .
Jakimovski and Leviatan [11], by virtue of the Appell polynomials A κ ( t ) , constructed a generalization of Szasz operators.
Under the following assumptions:
( i ) A κ ( t ) 0 , t [ 0 , ) ; ( i i ) A ( 1 ) 0 , ( i i i ) generating function ( 3 ) and the power series ( 4 ) converge for | x | < ( > 1 ) ,
The positive linear operators P η ( f ˇ ; t ) and their approximation properties are introduced by Jakimovski and Leviatan via
P η ( f ˇ ; t ) : = e η t A ( 1 ) κ = 0 A κ ( η t ) f ˇ κ η
After that, by taking the help of Sheffer polynomials s κ ( t ) , Ismail [9] introduced a generalization of Szasz and Jakimovski–Leviatan operators. The Sheffer sequences s κ ( t ) [12] are an important class of polynomial sequences and appear in various problems of applied mathematics, approximation theory, and multiple other mathematical areas.
The exponential generating function of s κ ( t ) is given by:
A ( x ) e t H ( x ) = κ = 0 s κ ( t ) x k ,
where
A ( x ) = κ = 0 a κ x κ , a 0 0 ,
H ( x ) = κ = 0 h κ x κ , h 1 0 .
Here we assume the following:
( i ) s κ ( t ) 0 , t [ 0 , ) ; ( i i ) A ( 1 ) 0 , H ( 1 ) = 1 , ( i i i ) g e n e r a t i n g f u n c t i o n ( 7 ) a n d p o w e r s e r i e s ( 8 ) a n d ( 9 ) c o n v e r g e f o r | x | < ( > 1 ) .
The properties of approximation of the positive linear operators investigated by Ismail [9] are given as
T η ( f ˇ ; t ) : = e η t H ( 1 ) A ( 1 ) κ = 0 s κ ( η t ) f ˇ κ η , η .
Additonally, Kantorovich investigated the approximation properties of positive linear operators defined by [13]
K η ( f ˇ ; t ) : = η e η t κ = 0 η ( η t ) κ κ ! κ / η ( κ + 1 ) / η f ˇ ( x ) d x .
Furthermore, many authors in [6,14,15,16] studied the approximation properties of the Szasz–Mirakyan–Kantorovich operators and their various propagation.
The polynomials defined in terms of the discrete convolution of known polynomials are applied to analyze new classes of special polynomials. Very recently, Khan and Riyasat [17] introduced and studied the discrete Appell convolution of the Sheffer polynomials s k ( x ) , called the Sheffer–Appell polynomials. These polynomials are denoted by s A κ ( t ) and are given with the help of the following generating function:
A ( x ) A ( x ) e t H ( x ) = κ = 0 s A κ ( t ) x κ ,
where A ( x ) , A ( x ) , and H ( x ) are given by Equation (4), Equation (8), and Equation (9), respectively.
It is important to note that the Sheffer–Appell polynomials are actually the Sheffer polynomials, since their generating function is of the type A * ( t ) e x H ( t ) , with a suitable choice for A * ( t ) . That means A * ( t ) is the product of two different functions of t, one of which corresponds to the Appell class, while the other should correspond to the Sheffer class.
In recent years, there is an increasing interest in modifying linear operators so that the new versions reproduce some basic functions, see for example [3,6,8,15,16]. Motivated by the works on generalizations of the Szasz operators, our aim is to construct the generalization of Szasz operators involving the Sheffer–Appell polynomials defined by generating function (13) and to study their approximation properties.
Since generating function (13) involves two different functions, A ( x ) , A ( x ) , and H ( x ) , therefore, we can define a new sequence of Kantrovich–Stancu-type approximation operators. It is important to note that the Sheffer–Appell polynomials are actually the Sheffer polynomials, since their generating function is of the type A * ( t ) e x H ( t ) , with a suitable choice for A * ( t ) . That means A * ( t ) is the product of two different functions of t, one of which corresponds to the Appell class, while the other should correspond to the Sheffer class.
The present work is organized as follows. In Section 2, the Kantrovich–Stancu-type positive linear operators together with the Sheffer–Appell polynomials are constructed, and their qualitative and quantitative results are derived. In Section 3, certain examples are given to demonstrate the results given in Section 2 with the help of the members belonging to the Sheffer–Appell polynomials.

2. Construction of Operators Y η ( δ , ϑ ) ( f ˇ ; t ) and Their Approximation Properties

In this section, a generalization of Kantrovich–Stancu operators is obtained with the help of Sheffer–Appell polynomials s A κ ( t ) . These operators are generalizations of Szasz operators (1) and Jakimovski–Leviatan operators (6). The convergence properties of these operators will also be established.
The operators are constructed together with the Sheffer–Appell polynomials s A κ ( t ) defined by Equation (13), under the following restrictions:
( i ) s A κ ( t ) 0 , κ = 0 , 1 , 2 , ; t [ 0 , ) , ( i i ) A ( 1 ) 0 , A ( 1 ) 0 , H ( 1 ) = 1 , ( i i i ) generating function ( 12 ) and power series ( 4 ) , ( 8 ) , and ( 9 ) converge for | x | < ( > 1 ) .
Now, under assumptions (14), a generalized form of positive linear operators as well as the Sheffer–Appell polynomials s A κ ( t ) is as follows:
Y η ( δ , ϑ ) ( f ˇ ; t ) = η + ϑ A ( 1 ) A ( 1 ) e η t H ( 1 ) κ = 0 s A κ ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ f ˇ x d x ,
where δ , ϑ are parameters satisfying 0 δ ϑ .
Here, note that for δ = ϑ = 0 , Y η ( δ , ϑ ) ( f ˇ ; t ) reduces (12).
Remark 1. 
For H ( x ) = x , (13) yields the generating function for the recently introduced 2-iterated Appell polynomials [18]
A ( x ) A ( x ) e t x = κ = 0 A κ [ 2 ] ( t ) x η .
Thus, taking H ( x ) = x in (15) and then denoting the resulting 2-iterated Appell polynomials in the right-hand side by A κ [ 2 ] ( η t ) , the following positive linear operators are obtained:
R η [ 2 ] ( δ , ϑ ) ( f ˇ ; t ) = η + ϑ A ( 1 ) A ( 1 ) e η t κ = 0 A κ [ 2 ] ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ f ˇ x d x .
Remark 2. 
For A ( x ) = 1 , (13) provides the generating function for the Sheffer polynomials. Therefore, taking A ( x ) = 1 in (15) we obtain Ismail operators (11), including the Sheffer polynomials.
Again, taking A ( x ) = 1 and H ( x ) = x in (15), we obtain Jakimovski and Leviatan operators (6) including the Appell polynomials.
Further, for A ( x ) = A ( x ) = 1 and H ( x ) = x , (15) reduces to Szasz operators (1).
Korovkin [1,2] demonstrated some noticeable outcomes concerning the convergence of sequences ( K η ( f ˇ , t ) ) η = 1 , where K η ( f ˇ , t ) , signify positive linear operators. For example, if K η ( f ˇ , t ) approaches uniformly to f ˇ in the specific cases viz. f ˇ ( x ) 1 , f ˇ ( x ) x , f ˇ ( x ) x 2 , then, at that point, it does likewise for each continuous real function f ˇ . Once more, if K η ( f ˇ , t ) approaches uniformly to f ˇ for specific cases f ˇ ( x ) 1 , cos x , sin x , then, at that point, it does likewise for each continuous 2 π periodic real function f ˇ . Shisha and Mond in [19] concluded the rate of convergence sequences K η ( f ˇ , t ) in terms of the moduli of continuity of f ˇ .
Our purpose is to obtain the theorem of convergence and the order of convergence of Y η ( δ , ϑ ) ( f ˇ ; t ) given by (15).
First, let us mention some useful definitions and results:
Definition 1. 
For any function f ˇ C ^ [ 0 , ) and σ > 0 , w ( f ˇ ; σ ) i.e., the modulus of continuity is defined by
w ( f ˇ ; σ ) : = sup u , v [ 0 , ) | u v | σ | f ˇ ( u ) f ˇ ( v ) | ,
where the space of uniformly continuous functions is given by C ^ [ 0 , ) . Note that for any σ > 0 and each u [ 0 , ) , one can write
| f ˇ ( u ) f ˇ ( v ) | w ( f ˇ ; σ ) | u v | σ + 1 .
Definition 2. 
For any function f ˇ C B [ 0 , ) , the second modulus of continuity is given by
w 2 ( f ˇ ; σ ) : = sup 0 < x σ | | f ˇ ( . + 2 x ) 2 f ˇ ( . + x ) + f ˇ ( . ) | | C B ,
where the family of real-valued bounded and uniformly continuous functions on [ 0 , ) is represented by C B [ 0 , ) , associated with the norm
| | f ˇ | | C B = sup t [ 0 , ) | f ˇ ( t ) | .
Definition 3. 
Let g ˇ C B [ 0 , ) , then the Peetre’s K-functional of g ˇ is given by
K ( g ˇ ; σ ) : = inf f ˇ C B 2 [ 0 , ) | | g ˇ f ˇ | | C B + σ | | f ˇ | | C B 2 ,
where
C B 2 [ 0 , ) : = { f ˇ C B [ 0 , ) : f ˇ , f ˇ C B [ 0 , ) }
and associated with the norm
| | f ˇ | | C B 2 : = | | f ˇ | | C B + | | f ˇ | | C B + | | f ˇ | | C B (see [20]).
Additionally, for all σ > 0 , the following inequality holds:
K ( g ˇ ; σ ) N { w 2 ( g ˇ , σ ) + m i n ( 1 , σ ) | | g ˇ | | C B } ,
where N is a constant independent of g ˇ and σ.
Lemma 1 
([7]). Suppose that we have the sequence of positive linear operators f ˇ C 2 [ 0 , a ] and ( K η ) η 0 with the property K η ( 1 ; t ) = 1 . Then,
| K η ( f ˇ ; t ) f ˇ ( t ) | | | f ˇ | | K η ( ( r t ) 2 ; t ) + 1 2 | | f ˇ | | K η ( ( r t ) 2 ; t ) .
Lemma 2 
([21]). Let f ˇ where f ˇ C [ c , d ] and ( 0 , c d 2 ) be the second-order Steklov function attached to f ˇ . Then, the inequalities
( a ) | | f ˇ f ˇ | | 3 4 w 2 ( f ˇ ; ) ,
( b ) | | f ˇ | | 3 2 2 w 2 ( f ˇ ; ) .
hold well.
In order to establish the theorem of convergence and the rate of convergence of Y η s A ( f ˇ , t ) including the Sheffer–Appell polynomials, the succeeding results are established:
Lemma 3. 
The Kantrovich–Stancu-type operators, defined by (15) are linear and positive.
Lemma 4. 
For t [ 0 , ) , we have the following properties of the operators Y η ( δ , ϑ ) ( f ˇ ; t )
Y η ( δ , ϑ ) ( 1 ; t ) = 1
Y η ( δ , ϑ ) ( s ; t ) = η η + ϑ t + 1 η + ϑ A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 δ + 1 2 ( η + ϑ ) Y η ( δ , ϑ ) ( s 2 ; t ) = η η + ϑ 2 t 2 + η ( η + ϑ ) 2 2 A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 ) + 2 ( δ + 1 ) + H ( 1 ) t + 1 ( η + ϑ ) 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 ) A ( 1 ) A ( 1 ) + δ 2 + δ + 1 3 ( η + ϑ ) 2
+ 2 ( δ + 1 ) ( η + ϑ ) 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) .
Proof. 
From generating function (13), it follows that
κ = 0 s A κ ( η t ) = A ( 1 ) A ( 1 ) e η t H ( 1 ) ,
κ = 0 κ s A κ ( η t ) = A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + η t A ( 1 ) A ( 1 ) e η t H ( 1 ) , κ = 0 κ 2 s A κ ( η t ) = η 2 t 2 A ( 1 ) A ( 1 ) + η t H ( 1 ) A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 )
+ A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) e η t H ( 1 ) .
Using operator Equation (15) and generating function (13) in Equations (28)–(30), assertions (25)–(27) are obtained. □
Lemma 5. 
For t [ 0 , ) , the following identities holds for the operators Y η ( δ , ϑ ) ( f ˇ ; t )
Y η ( δ , ϑ ) ( s t ; t ) = ϑ η + ϑ t + 1 η + ϑ A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 δ + 1 2 ( η + ϑ ) Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) = ϑ n + ϑ 2 t 2 + η ( n + ϑ ) 2 { 2 ϑ A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 ( δ + 1 ) + H ( 1 ) ( δ + 1 / 2 ) ( η + ϑ η ) } t + 2 ( δ + 1 ) ( n + ϑ ) 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 )
+ 1 ( η + ϑ ) 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 ) A ( 1 ) A ( 1 ) + δ 2 + δ + 1 3 ( η + ϑ ) 2 .
Proof. 
In view of the linearity property of Y η ( δ , ϑ ) , it follows that
Y η ( δ , ϑ ) ( s t ; t ) = Y η ( δ , ϑ ) ( s ; t ) t Y η ( δ , ϑ ) ( 1 ; t )
Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) = Y η ( δ , ϑ ) ( s 2 ; t ) 2 t Y η ( δ , ϑ ) ( s ; t ) + t 2 Y η ( δ , ϑ ) ( 1 ; t ) ,
which on applying Lemma 4, yields assertions (36) and (37), respectively. □
We denote the set of all continuous functions by C E [ 0 , ) , such that | f ˇ ( t ) | γ ρ t for all t 0 and for some positive finite ρ and γ .
The theorem of convergence for the operators Y η ( δ , ϑ ) ( f ˇ , t ) is obtained by proving the following result:
Theorem 1. 
Let f ˇ C E [ 0 , ) . Then
lim n Y η ( δ , ϑ ) ( f ˇ , t ) = f ˇ ( t )
uniformly for every compact subset of [ 0 , ) .
Proof. 
With the aid of Lemma 3, we get
lim η Y η ( δ , ϑ ) ( s i , t ) = t i , i = 0 , 1 , 2
uniformly for every compact subset of [ 0 , ) . Applying Korovkin’s theorem to (36) proves the desired result. □
Next, we determine the convergence rate of Y η ( δ , ϑ ) ( f ˇ , t ) by using the modulus of continuity in the form of the following result:
Theorem 2. 
Suppose f ˇ C ^ E [ 0 , ) . Then the following inequality
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 2 w ( f ˇ ; μ η ( t ) ,
holds for Y η ( δ , ϑ ) ( f ˇ , t ) . Here,
μ η ( x ) : = ( Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) .
Proof. 
Consider
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ( x ) | = | η + ϑ A ( 1 ) A ( 1 ) e η t H ( 1 ) κ = 0 s A κ ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ f κ η f ˇ ( t ) d t |
Using the fact that Y η ( δ , ϑ ) ( 1 ; t ) = 1 and in view of (15), we have
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | η + ϑ A ( 1 ) A ( 1 ) e η t H ( 1 ) κ = 0 s A κ ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ f ˇ κ η f ˇ ( t ) d t ,
which on using Equation (19) gives
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 1 + η + ϑ A ( 1 ) A ( 1 ) e η t H ( 1 ) 1 σ κ = 0 s A κ ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ κ η t w ( f ˇ ; δ ) d t .
By considering the Cauchy-Schwarz inequality for integration, we find
κ + δ η + ϑ κ + δ + 1 η + ϑ κ η t d t 1 η + ϑ κ + δ η + ϑ κ + δ + 1 η + ϑ κ η t 2 1 2 ,
from this, it follows that
κ = 0 s A κ ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ κ η t d t 1 η + ϑ κ = 0 s A κ ( n t ) κ + δ η + ϑ κ + δ + 1 η + ϑ κ η t 2 1 2 .
Using Cauchy-Schwarz inequality for summation on the right-hand side of (44), we can write
κ = 0 s A κ ( η t ) κ + δ η + ϑ κ + δ + 1 η + ϑ κ η t d t A ( 1 ) A ( 1 ) e η t H ( 1 ) η + ϑ A ( 1 ) A ( 1 ) e η t H ( 1 ) η + ϑ Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) 2 1 2 .
From the inequality in (42), we find that
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 1 + 1 σ μ η ( t ) w ( f ˇ ; δ ) .
By taking σ = μ η ( t ) in above equation, the assertion in Equation (40) is established. □
Theorem 3. 
For f ˇ C [ 0 , δ ] , the succeeding inequality holds:
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 2 δ | | f ˇ | | l 2 + 3 4 ( δ + 2 + l 2 ) w 2 ( f ˇ ; l ) ,
where
l : = l η ( t ) = Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) 1 4
and w 2 ( f ˇ ; l ) is the second order modulus of continuity with norm | | f ˇ | | = max t [ a , b ] | f ˇ ( t ) |
Proof. 
Let f ˇ l be the second-order Steklov function of the function f. So, in view of (28), we have
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | | Y η ( δ , ϑ ) ( f ˇ f ˇ l ; t ) | + | Y η ( δ , ϑ ) ( f ˇ l ; t ) f ˇ l ( t ) | + | f ˇ l ( t ) f ˇ ( t ) | , 2 | | f ˇ l f ˇ | | + | Y η ( δ , ϑ ) ( f ˇ l ; t ) f ˇ l ( t ) | ,
on using (26), (48) becomes
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 3 2 w 2 ( f ˇ ; l ) + | Y η ( δ , ϑ ) ( f ˇ l ; t ) f ˇ l ( t ) | .
Keeping in view that f ˇ l C 2 [ 0 , δ ] , from Lemma 2, it follows that
| Y η ( δ , ϑ ) ( f ˇ l ; t ) f ˇ l ( t ) | | | f ˇ l | | Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) + 1 2 | | f ˇ l | | Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) ,
now on on using (27), (50) becomes
| Y η ( δ , ϑ ) ( f ˇ l ; t ) f ˇ l ( t ) | | | f ˇ l | | Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) + 3 4 l 2 w 2 ( f ˇ ; l ) Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) .
Further, the Landau inequality
| | f ˇ l | | 2 δ | | f ˇ l | | + δ 2 | | f ˇ l | | ,
combined with inequality (27) gives
| | f ˇ l | | 2 δ | | f ˇ | | + 3 δ 4 l 2 w 2 ( f ˇ ; l ) .
Substituting (52) in (53) and taking l = Y η ( δ , ϑ ) ( ( s t ) 2 ) ; t 4 , we find
| Y η ( δ , ϑ ) ( f ˇ l ; t ) f ˇ l ( t ) | 2 δ | | f ˇ | | l 2 + 3 4 ( δ + l 2 ) w 2 ( f ˇ ; l ) .
Making use of (48) in (44), we arrive at (42). □
Theorem 4. 
Let f ˇ C B 2 [ 0 , ) . If Y η ( δ , ϑ ) is defined by (15), then one has
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | ζ | | v | | C B 2 ,
where
ζ = ζ η ( t ) = [ ϑ η + ϑ 2 t 2 + η ( η + ϑ ) 2 { 2 δ + 2 + H ( 1 ) 2 ϑ A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) δ + 1 / 2 + η ϑ ( η + ϑ ) } t + 2 δ + 2 + η + ϑ ( η + ϑ ) 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 δ + 1 2 ( η + ϑ ) + 1 ( η + ϑ ) 2 A ( 1 ) A ( 1 ) + A ( 1 ) A ( 1 ) + 2 A ( 1 ) A ( 1 ) A ( 1 ) A ( 1 ) + δ 2 + δ + 1 3 ( η + ϑ ) 2 ] | | f ˇ | | C B 2 .
Proof. 
In view of the linearity property of Y η ( δ , ϑ ) ( f ˇ ; t ) and Y η ( δ , ϑ ) ( 1 ; t ) = 1 , the Taylor’s expansion of function f ˇ can be written as:
Y η ( δ , ϑ ) ( f ˇ ; t ) = f ( x ) Y η ( δ , ϑ ) ( 1 ; t ) + f ˇ ( t ) Y η ( δ , ϑ ) ( s t ; t ) + 1 2 f ˇ ( ξ ) Y η ( δ , ϑ ) ( ( s t ) 2 ; t ) , ξ ( t , s ) .
From Lemma 5 Y η ( δ , ϑ ) ( s t ; t ) 0 for s t. Thus by inserting expressions (36) and (37) in (56) and after simplification we get the required result. □
Theorem 5. 
If f ˇ is a function such that f ˇ C B [ 0 , ) , then
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 2 N { w 2 ( f ˇ ; σ ) + m i n ( 1 , σ ) | | f ˇ | | C B } ,
where
σ : = σ η ( t ) = 1 2 ζ η ( t ) ,
where N 0 is a constant and is independent of f ˇ , δ. Additionally, ζ η ( t ) is given in Theorem 4.
Proof. 
If g ˇ C B 2 [ 0 , ) , then from Theorem 4, we have
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | | Y η ( δ , ϑ ) ( f ˇ g ˇ ; t ) | + | Y η ( δ , ϑ ) ( g ˇ ; t ) g ˇ ( t ) | + | g ˇ ( t ) f ˇ ( t ) | 2 | | f ˇ g ˇ | | C B + ξ | | g ˇ | | C B 2 = 2 | | f ˇ g ˇ | | C B + δ | | g ˇ | | C B 2
Since the left-hand side of (59) is independent of the function g ˇ C B 2 [ 0 , ) ,
| Y η ( δ , ϑ ) ( f ˇ ; t ) f ˇ ( t ) | 2 K ( f ˇ ; δ ) ,
where K is Peetre’s functional defined by (22). Now, by using (24) in (59), (57) holds. □
Remark 3. 
In Theorems 3–5, l n , ξ n , δ n 0 when n .
In the next section, we consider certain examples in support of the above-derived results.

3. Examples

We establish the positive linear operators including certain members of the Sheffer–Appell family by considering the following examples.
Example 1. 
The truncated exponential polynomials e κ ( t ) [22] are the Appell polynomials for A ( x ) = 1 1 x and are defined by the generating function:
1 ( 1 x ) e t x = κ = 0 e κ ( t ) x κ .
These polynomials have many applications in optics and quantum mechanics and also perform a key role in the evaluation of integrals having products of special functions.
The Laguerre polynomials L κ ( t ) [23] are essential members of the Sheffer family for A ( x ) = 1 1 x and H ( x ) = x 1 x and are defined by the following generating function:
1 ( 1 x ) e t x 1 x = κ = 0 L κ ( t ) x κ .
The Laguerre polynomials arise in the quantum mechanics of the Morse potential and of the 3D isotropic consonant oscillator, in an outspread piece of the arrangement of the Schrödinger condition for a one-electron iota.
Taking A ( x ) = 1 1 x of the truncated exponential polynomials and A ( x ) = 1 1 x ; H ( x ) = x 1 x of Laguerre polynomials in generating function (12), the following generating function of the Laguerre-truncated exponential polynomials L e κ ( t ) is obtained:
1 ( 1 x ) 2 e ( t x 1 x ) = κ = 0 L e κ ( t ) x κ .
For ensuring restrictions (14), generating function (63) is modified by replacing x x 2 and t t 2 as follows:
1 1 x 2 2 e t x 2 ( 2 x ) = κ = 0 L e κ ( t / 2 ) x κ 2 κ ,
which yields the following explicit representation of L e κ ( t / 2 ) :
L e κ ( t / 2 ) = l , s = 0 l + s κ ( l + 1 ) ( κ l s ) s t κ l s 2 κ l s ( κ l s ) ! s ! .
From Equation (65), it follows that
L e κ ( t / 2 ) 0 for all t [ 0 , ) .
In view of generating function (64), the positive linear operators including the Laguerre-truncated exponential polynomials are constructed as follows:
Y η L e ( δ , ϑ ) ( f ˇ ; t ) = 1 4 e η t / 2 κ = 0 κ + δ η + ϑ κ + δ + 1 η + ϑ L e κ ( η t / 2 ) f ˇ κ η d t .
Example 2. 
The polynomials denoted by g κ d + 1 ( t ; h ) [24] are called Gould–Hopper d-orthogonal polynomial sets [25,26] of Hermite type [14]. For A ( x ) = e h x d + 1 these polynomials are the Appell polynomials given by the generating relation:
e h x d + 1 e t x = κ = 0 g κ d + 1 ( x ; h ) x κ .
Taking A ( x ) = e h x d + 1 of the Gould–Hopper polynomials and A ( x ) = 1 1 x ; H ( x ) = x 1 x of Laguerre polynomials in generating function (12), the following generating function for the Laguerre–Gould–Hopper polynomials L g κ d + 1 ( t ; h ) is obtained:
1 ( 1 x ) e h x d + 1 e ( t x 1 x ) = κ = 0 L g κ d + 1 ( t ; h ) x k .
For ensuring restrictions (14), generating function (68) is modified by replacing t x 2 and t t 2 as follows:
1 ( 1 x 2 ) e h ( x 2 ) d + 1 e t x 2 ( 2 x ) = κ = 0 L g κ d + 1 ( t / 2 ; h ) x κ 2 κ .
From Equation (69), the following explicit representation of L g κ d + 1 ( t / 2 ; h ) is obtained:
L g κ d + 1 ( t / 2 ; h ) = l , s , v = 0 l + ( d + 1 ) s + v κ h s ( κ l ( d + 1 ) s v ) v t κ l ( d + 1 ) s v 2 κ l ( d + 1 ) s v ( κ l ( d + 1 ) s v ) ! s ! v ! .
Again, from Equation (70), it follows that
L g κ d + 1 ( t / 2 ; h ) 0 for all t [ 0 , ) .
In view of generating function (69), the positive linear operators together with the Laguerre–Gould–Hopper polynomials are constructed as follows:
Y η L g d + 1 ( δ , ϑ ) ( f ˇ ; t ) = 1 2 e 1 2 η t + h 2 d κ = 0 κ + δ η + ϑ κ + δ + 1 η + ϑ L e κ ( η t / 2 ) f ˇ κ η d t .
The methodology adopted above can be extended to find the positive linear operators including other members of the Sheffer–Appell family, provided these polynomials obey restrictions (14).

4. Concluding Remarks

In this article, the positive linear operators together with the Sheffer–Appell polynomials are introduced. The convergence theorem and rate of convergence of these operators are also established. The Sheffer–Appell family includes a large number of hybrid-type polynomials as its members. Some examples are also provided to give the importance of the operators including the Sheffer–Appell polynomials.
The error estimation for the approximation with the operators including members of the Sheffer–Appell family can be explored. The approximation of any continuous function f ˇ ( t ) by positive linear operators Y η s A ( f ˇ ; t ) can also be shown graphically. Finding the Kantrovich, Durrmeyer, and Kantrovich–Stancu-type generalizations of the operators including Sheffer–Appell polynomials will be taken in a forthcoming investigation. This article is a first attempt in the direction of finding generalizations of the Szasz operators involving hybrid-type families of special functions.

Author Contributions

Both authors have contributed equally to this manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Not available.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Korovkin, P.P. On convergence of linear positive operators in the space of continous functions. Dokl. Akad. Nauk SSSR 1953, 90, 961–964. [Google Scholar]
  2. Korovkin, P.P. Linear Operators and Approximation Theory; Chapters I, II. (Russian, 1959) English Translation; Hindustan Publishing Corp.: Delhi, India, 1960. [Google Scholar]
  3. Sucu, S.; z, G.; Varma, S. On some extensions of Szasz operators including Boas-Buck-type polynomials. Abstr. Appl. Anal. 2012, 1–15. [Google Scholar] [CrossRef] [Green Version]
  4. Mursaleen, M.; Ansari, K.J.; Khan, A. On (p,q)-analogue of Bernstein Operators. Appl. Math. Comput. 2015, 266, 874–882. [Google Scholar] [CrossRef]
  5. Szasz, O. Generalization of S. Bernstein’s polynomials to the infinite interval. J. Res. Nat. Bur. Stand. 1950, 45, 239. [Google Scholar] [CrossRef]
  6. Varma, S.; Sucu, S.; z, G. Generalization of Szasz operators involving Brenke type polynomials. Comput. Math. Appl. 2012, 64, 121–127. [Google Scholar] [CrossRef] [Green Version]
  7. Gavrea, I.; Raşa, I. Remarks on some quantitative Korovkin-type results. Rev. Anal. Numér. Théor. Approx. 1993, 22, 173–176. [Google Scholar]
  8. z, G.; Varma, S.; Sucu, S. Approximation by operators including generalized Appell polynomials. Filomat 2016, 30, 429–440. [Google Scholar]
  9. Ismail, M.E.H. On a generalization of Szasz operators. Mathematica (Cluj) 1974, 39, 259–267. [Google Scholar]
  10. Appell, P. Sur une classe de polynômes. Ann. Sci. École. Norm. Sup. 1880, 9, 119–144. [Google Scholar] [CrossRef] [Green Version]
  11. Jakimovski, A.; Leviatan, D. Generalized Szasz operators for the approximation in the infinite interval. Mathematica (Cluj) 1969, 11, 97–103. [Google Scholar]
  12. Sheffer, I.M. Some properties of polynomial sets of type zero. Duke. Math. J. 1939, 5, 590–622. [Google Scholar] [CrossRef]
  13. Butzer, P.L. On the extensions of Bernstein polynomials to the infinite interval. Proc. Am. Math. Study 1954, 5, 547–553. [Google Scholar] [CrossRef]
  14. Douak, K. The relation of the d-orthogonal polynomials to the Appell polynomials. J. Comput. Appl. Math. 1996, 70, 279–295. [Google Scholar] [CrossRef] [Green Version]
  15. Alyusof, R.; Jeelani, M.B. Some Families of differential equations associated with the Gould-Hopper-Frobenius-Genocchi polynomials. Aims Math. 2022, 7, 4851–4860. [Google Scholar] [CrossRef]
  16. Jeelani, M.B. On employing linear algebra approach to hybrid Sheffer polynomials. Aims Math. 2023, 8, 1871–1888. [Google Scholar] [CrossRef]
  17. Khan, S.; Riyasat, M. A determinantal approach to Sheffer-Appell polynomials via monomiality principle. J. Math. Anal. App. 2015, 421, 806–829. [Google Scholar] [CrossRef]
  18. Khan, S.; Raza, N. 2-Iterated Appell polynomials and related numbers. Appl. Math. Comput. 2013, 219, 9469–9483. [Google Scholar] [CrossRef]
  19. Shisha, O.; Mond, B. The degree of convergence of linear positive operators. Proc. Natl. Acad. Sci. USA 1968, 60, 1196–1200. [Google Scholar] [CrossRef] [Green Version]
  20. Ditzian, Z.; Totik, V. Moduli of Smoothness; Vol.9 of Springer Series in Computational Mathematics; Springer: New York, NY, USA, 1987. [Google Scholar]
  21. Zhuk, V.V. Functions of the Lip 1 class and S. N. Bernstein’s polynomials. Vestnik Leningrad. Univ. Mat. Mekh. Astronom. 1989, 1, 25–30. [Google Scholar]
  22. Dattoli, G.; Cesarano, C.; Sacchetti, D. A note on truncated polynomials. Appl. Math. Comput. 2003, 134, 595–605. [Google Scholar] [CrossRef]
  23. Andrews, L.C. Special Functions for Engineers and Applied Mathematicians; Macmillan Publishing Company: New York, NY, USA, 1985. [Google Scholar]
  24. Gould, H.W.; Hopper, A.T. Operational formulas connected with two generalizations of Hermite polynomials. Duke. Math. J. 1962, 29, 51–63. [Google Scholar] [CrossRef]
  25. Iseghem, J.V. Vector orthogonal relations. Vector QD-algorithm. J. Comput. App. Math. 1987, 19, 141–150. [Google Scholar] [CrossRef] [Green Version]
  26. Maroni, P. L’orthogonalité et les récurrences de polynômes d’ordre supérieur à deux. Ann. Fac. Sci. Toulouse Math. 1989, 10, 105–139. [Google Scholar] [CrossRef]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Jeelani, M.B.; Alnahdi, A.S. Approximation by Operators for the Sheffer–Appell Polynomials. Symmetry 2022, 14, 2672. https://doi.org/10.3390/sym14122672

AMA Style

Jeelani MB, Alnahdi AS. Approximation by Operators for the Sheffer–Appell Polynomials. Symmetry. 2022; 14(12):2672. https://doi.org/10.3390/sym14122672

Chicago/Turabian Style

Jeelani, Mdi Begum, and Abeer S. Alnahdi. 2022. "Approximation by Operators for the Sheffer–Appell Polynomials" Symmetry 14, no. 12: 2672. https://doi.org/10.3390/sym14122672

APA Style

Jeelani, M. B., & Alnahdi, A. S. (2022). Approximation by Operators for the Sheffer–Appell Polynomials. Symmetry, 14(12), 2672. https://doi.org/10.3390/sym14122672

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop