• 検索結果がありません。

SOME RESULTS FOR THE APOSTOL-GENOCCHI POLYNOMIALS OF HIGHER ORDER

N/A
N/A
Protected

Academic year: 2022

シェア "SOME RESULTS FOR THE APOSTOL-GENOCCHI POLYNOMIALS OF HIGHER ORDER"

Copied!
18
0
0

読み込み中.... (全文を見る)

全文

(1)

POLYNOMIALS OF HIGHER ORDER

1∗HASSAN JOLANY,HESAM SHARIFI AND R. EIZADI ALIKELAYE

1*School of Mathematics, Statistics and Computer Science, University of Tehran, Iran. E-mail: [email protected]

Department of Mathematics, Faculty of Science, University of Shahed, Tehran, Iran. E-mail: [email protected]

Abstract. The present paper deals with multiplication formulas for the Apostol-Genocchi polynomials of higher order and deduces some ex- plicit recursive formulas. Some earlier results of Carlitz and Howard in terms of Genocchi numbers can be deduced. We introduce the 2-variable Apostol-Genocchi polynomials and then we consider the multiplication theorem for 2-variable Genocchi polynomials. Also we introduce gen- eralized Apostol-Genocchi polynomials with a, b, cparameters and we obtain several identities on generalized Apostol-Genocchi polynomials witha, b, cparameters .

Keywords and Phrases:Apostol-Genocchi numbers and polynomials (of higher order), Generalization of Genocchi numbers and polynomials, Raabe’s multiplication formula, multiplication formula, Bernoulli num- bers and polynomials, Euler numbers and polynomials, Stirling numbers

1. Preliminaries and motivation

The classical Genocchi numbers can be defined in a number of ways. The way in which it is defined is often determined by which sorts of applications they are intended to be used for. The Genocchi numbers have wide-ranging applications from number theory and Combinatorics to numerical analysis and other fields of applied mathematics. There exist two important defini- tions of the Genocchi numbers: the generating function definition, which is the most commonly used definition, and a Pascal-type triangle definition, first given by Philip Ludwig von Seidel, and discussed in [29]. As such, it makes it very appealing for use in combinatorial applications. The idea behind this definition, as in Pascal’s triangle, is to utilize a recursive relation- ship giving some initial conditions to generate the Genocchi numbers. The combinatorics of the Genocchi numbers were developed by Dumont in [4] and various co-authors in the 70s and 80s. Dumont and Foata introduced in 1976

2010Mathematics Subject Classification. 11B68, 05A10, 05A15.

1corresponding Author

1

(2)

a three-variable symmetric refinement of Genocchi numbers, which satisfies a simple recurrence relation. A six-variable generalization with many simi- lar properties was later considered by Dumont. In [30] Jang et al. defined a new generalization of Genocchi numbers, poly Genocchi numbers. Kim in [10] gave a new concept for the q-extension of Genocchi numbers and gave some relations between q-Genocchi polynomials and q-Euler numbers. In [31], Simsek et al. investigated the q-Genocchi zeta function and L-function by using generating functions and Mellin transformation. Genocchi numbers are known to count a large variety of combinatorial objects, among which numerous sets of permutations. One of the applications of Genocchi num- bers that was investigated by Jeff Remmel in [32] is counting the number of up-down ascent sequences. Another application of Genocchi numbers is in Graph Theory. For instance, Boolean numbers of the associated Ferrers Graphs are the Genocchi numbers of the second kind [33]. A third applica- tion of Genocchi numbers is in Automata Theory. One of the generalizations of Genocchi numbers that was first proposed by Han in [34] proves useful in enumerating the class of deterministic finite automata (DFA) that accept a finite language and in enumerating a generalization of permutations counted by Dumont. Recently S. Herrmann in [6], presented a relation between the f -vector of the boundary and the interior of a simplicial ball directly in terms of the f-vectors. The most interesting point about this equation is the occurrence of the Genocchi numbersG2n. In the last decade, a surpris- ing number of papers appeared proposing new generalizations of the classical Genocchi polynomials to real and complex variables or treating other top- ics related to Genocchi polynomials. Qiu-Ming Luo in [19] introduced new generalizations of Genocchi polynomials, he defined the Apostol-Genocchi polynomials of higher order and q-Apostol-Genocchi polynomials and he ob- tained a relationship between Apostol-Genocchi polynomials of higher order and Goyal-Laddha-Hurwitz-Lerch Zeta function. Next Qiu-Ming Luo and H.M. Srivastava in [35] by Apostol-Genocchi polynomials of higher order derived various explicit series representations in terms of the Gaussian hy- pergeometric function and the Hurwitz (or generalized) zeta function which yields a deeper insight into the effectiveness of this type of generalization.

Also it is clear that Apostol-Genocchi polynomials of higher order are in a class of orthogonal polynomials and we know that most such special func- tions that are orthogonal are satisfied in multiplication theorem, so in this present paper we show this property is true for Apostol-Genocchi polyno- mials of higher order.

The study of Genocchi numbers and their combinatorial relations has received much attention [2, 4, 6, 10, 13, 19, 22, 23, 26, 27, 29, 37]. In this paper we consider some combinatorial relationships of the Apostol-Genocchi numbers of higher order.

The unsigned Genocchi numbers{G2n}n>1 can be defined through their generating function:

(3)

X

n=1

G2n x2n

(2n)! =x.tanx 2

and also

X

n>1

(−1)nG2n t2n

(2n)! =−ttanht 2

So, by simple computation tanh

t 2

= X

s>0

(2t)2s+1 (2s+ 1)!.X

m>0

(−1)mE2m

(2t)2m (2m)!

= X

s,m>0

(−1)m 22m+2s+1

E2mt2m+2s+1 (2m)!(2s+ 1)!

= X

n>1 n−1

X

m=0

2n−1 2m

(−1)mE2mt2n−1 22n−1(2n−1)! , we obtain forn>1,

G2n=

n−1

X

k=0

(−1)n−k−1(n−k) 2n

2k E2k

22n−2

whereEkare Euler numbers. Also the Genocchi numbersGnare defined by the generating function

G(t) = 2t et+ 1 =

X

n=0

Gntn

n!,(|t|< π).

In general, it satisfies G0 = 0, G1 = 1, G3 = G5 = G7 = ...G2n+1 = 0, and even coefficients are given G2n = 2(1−22n)B2n = 2nE2n−1, where Bn are Bernoulli numbers and En are Euler numbers. The first few Genocchi numbers for even integers are -1,1,-3,17,-155,2073,... . The first few prime Genocchi numbers are -3 and 17, which occur atn= 6 and 8. There are no others with n <105. Forx ∈ R, we consider the Genocchi polynomials as follows

G(x, t) =G(t)ext= 2t

et+ 1ext=

X

n=0

Gn(x)tn n!. In special casex= 0, we define Gn(0) =Gn. Because we have

Gn(x) =

n

X

k=0

n k

Gkxn−k,

It is easy to deduce thatGk(x) are polynomials of degreek. Here, we present some of the first Genocchi’s polynomials:

G1(x) = 1, G2(x) = 2x−1, G3(x) = 3x2−3x, G4(x) = 4x3−6x2+ 1,

(4)

G5(x) = 5x4−10x3+ 5x, G6(x) = 6x5−15x4+ 15x2−3, ...

The classical Bernoulli polynomials (of higher order)B(α)n (x) and Euler poly- nomials (of higher order) En(α)(x),(α ∈C), are usually defined by means of the following generating functions [11, 13, 15, 21, 24, 25, 28]

z ez−1

α

exz =

X

n=0

Bn(α)(x)zn

n!,(|z|<2π) and

2 ez+ 1

α

exz =

X

n=0

En(α)(x)zn

n!,(|z|< π) So that, obviously,

Bn(x) :=Bn1(x) and En(x) :=En(1)(x).

In 2002, Q. M. Luo et al. (see [5, 17, 18]) defined the generalization of Bernoulli polynomials and Euler numbers, as follows

tcxt bt−at =

X

n=0

Bn(x;a, b, c)

n! tn,(|tlnb

a|<2π) 2

bt+at =

X

n=0

En(a, b)tn

n!,(|tlnb a|< π).

Here, we give an analogous definition for generalized Apostol-Genocchi poly- nomials.

Let a, b > 0, The Generalized Apostol-Genocchi Numbers and Apostol- Genocchi polynomials witha, b, c parameters are defined by

2t λbt+at =

X

n=0

Gn(a, b;λ)tn n!

2t

λbt+atext=

X

n=0

Gn(x, a, b;λ)tn n!

2t

λbt+atcxt=

X

n=0

Gn(x, a, b, c;λ)tn n!

respectively.

For a real or complex parameter α, The Apostol-Genocchi polynomials with a, b, c parameters of order α, G(α)n (x;a, b;λ), each of degree n is x as well as inα, are defined by the following generating functions

2t λbt+at

α

exz =

X

n=0

G(α)n (x, a, b;λ)tn n!, Clearly, we have G(1)n (x, a, b;λ) =Gn(x;a, b;λ).

(5)

Now, we introduce the 2-variable Apostol-Genocchi polynomials and then we consider the multiplication theorem for 2-variable Apostol-Genocchi Poly- nomials.

We start with the definition of Apostol-Genocchi polynomials Gn(x;λ).

The Apostol-Genocchi Polynomials Gn(x;λ) in variable x are defined by means of the generating function

2zexz λez+ 1 =

X

n=0

Gn(x;λ)zn

n! (|z|<2π whenλ= 1,|z|<|logλ|whenλ6= 1), with, of course,

Gn(λ) :=Gn(0;λ),

WhereGn(λ) denotes the so-called Apostol-Genocchi numbers.

Also (see [1, 14, 16, 19, 20, 24, 28]) Apostol-Genocchi PolynomialsG(α)n (x;λ) of orderα in variable x are defined by means of the generating function:

2z λez+ 1

α

exz =

X

n=0

G(α)n (x;λ)zn n!

with, of course,G(α)n (λ) :=Gαn(0;λ).

Where Gαn(λ) denotes the so-called Apostol-Genocchi numbers of higher order. If we set,

φ(x, t;α) = 2t et+ 1

α

ext, then,

∂φ

∂x =tφ, and,

t∂φ

∂t −nα+tx

t − αet et+ 1

o∂φ

∂x = 0.

Next, we introduce the class of Apostol-Genocchi numbers as follows. (for more information see [38])

HGn(λ) =

[n2]

X

s=0

n!Gn−2s(λ)Gs(λ) s!(n−2s)!

The generating function ofHGn(λ) is provided by 4t3

(λet+ 1)(λet2 + 1) =

X

n=0

HGn(λ)tn n!

and the generalization ofHGn(λ) for (a, b)6= 0, is 4t3

(λeat+ 1)(λebt2 + 1) =

X

n=0

HGn(a, b;λ)tn n!

(6)

where

HGn(a, b;λ) = 1 ab

[n2]

X

n=0

n!an−2sbsGn−2s(λ)Gs(λ) s!(n−2s)!

The main object of the present paper is to investigate the multiplication formulas for the Apostol-type polynomials.

Luo in [16] defined the multiple alternating sums as Zk(l)(m;λ) = (−1)l X

0≤v1,v2,...,vm≤l v1+v2+...+vm=`

l v1, v2, ..., vm

(−λ)v1+2v2+...+mvm

Zk(m;λ) =

m

X

j=1

(−1)j+1λjjk=λ−λ22k+...+ (−1)m+1λmmk

Zk(m) =

m

X

j=1

(−1)j+1jk= 1−2k+...+ (−1)m+1mk, (m, k, l∈N0;λ∈C) whereN0 :=N∪ {0} , (N:={1,2,3, ...}).

2. the multiplication formulas for the apostol-genocchi polynomials of higher order

In this Section, we obtain some interesting new relations and properties associated with Apostol-Genocchi polynomials of higher order and then de- rive several elementary properties including recurrence relations for Genoc- chi numbers. First of all we prove the multiplication theorem of these poly- nomials.

Theorem 2.1. For m ∈N, n ∈N0, α, λ ∈ C, the following multiplication formula of the Apostol-Genocchi polynomials of higher order holds true:

(1)

G(α)n (mx;λ) =mn−α X

v1,v2,...,vm−1≥0

α v1, v2, ..., vm−1

(−λ)rG(α)n x+ r

m;λm where r =v1+ 2v2+...+ (m−1)vm−1, (m is odd )

Proof. It is easy to observe that 1

λet+ 1 =−1−λet2e2t+...+ (−λ)m−1e(m−1)t

(−λ)memt−1 (∗)

But we have, if xi ∈C

(x1+x2+...+xm)n= X

a1,a2,...,am>0 a1+a2+...am=n

n a1, a2, ..., am

xa11xa22...xamm (∗∗)

(7)

The last summation takes place over all positive or zero integersai>0 such thata1+a2+...+am =n, where

n a1, a2, ..., am

:= n!

a1!a2!...am!

So by applying (∗) on the following first equality sign and setting (x1 = 1, xk = (−λ)kekt for k ≥ 2) and n = α in (∗∗) on the following second equality sign, we obtain

X

n=0

G(α)n (mx;λ)tn

n! = 2t λet+ 1

α

emxt

=

2t λmemt+ 1

αm−1X

k=0

(−λ)kekt α

emxt

= X

v1,v2,...,vm−1>0

α v1, v2, ..., vm−1

(−λ)r 2t λmemt+ 1

α

e(x+mr)mt

=

X

n=0

mn−α X

v1,v2,...,vm>0

α v1, v2, ..., vm

(−λ)rG(α)n x+ r

m;λm tn n!

By comparing the coefficient of tn!n on both sides of last equation, proof is

complete.

In terms of the generalized Apostol-Genocchi polynomials, by settingλ= 1 in Theorem 2.1, we obtain the following explicit formula that is called multiplication theorem for Genocchi polynomials of higher order.

Corollary 2.2. Form∈N, n∈N0, α,∈C, we have

G(α)n (mx) =mn−α X

v1,v2,...,vm−1>0

α v1, v2, ..., vm−1

(−1)rG(α)n

x+r m

(m is odd).

And using Corollary 2.2, (by setting α = 1), we get Corollary 2.3 that is the main result of [36] and is called multiplication Theorem for Genocchi polynomials.

Corollary 2.3. Form∈N, n∈N0, we have

Gn(mx) =mn−1

m−1

X

k=0

(−1)kGn

x+ k

m

(m is odd).

Now, we consider the multiplication formula for the Apostol-Genocchi numbers when m is even.

Theorem 2.4. For m ∈ N (m even), n ∈ N, α, λ ∈ C, the following multiplication formula of the Apostol-Genocchi polynomials of higher order

(8)

holds true:

G(α)n (mx;λ) = (−2)αmn−α X

v1,v2,...,vm−1>0

α v1, v2, ..., vm−1

(−λ)rBn(α)

x+r m, λm

, where r =v1+ 2v2+...+ (m−1)vm−1.

Proof. It is easy to observe that 1

λet+ 1 =−1−λet2e2t+...+ (−λ)m−1e(m−1)t (−λ)memt−1

So, we obtain

X

n=0

G(α)n (mx;λ)tn

n! = 2t λet+ 1

α

emxt

= 2α t

λet+ 1 α

emxt

= (−2)α t λmemt−1

αm−1X

k=0

(−λet)kα

emxt

= (−2)α X

v1,v2,...,vm−1>0

α v1, v2, ..., vm−1

(−λ)r t λmem−1

α

e(x+mr)mt

=

X

n=0

(−2)αmn−α X

v1,v2,...,vm−1>0

α v1, v2, ..., vm−1

(−λ)r

× B(α)n (x+ r

m;λm)tn n!

By comparing the coefficients of tn!n on both sides proof will be complete.

Next, using Theorem 2.4, (with λ = 1), we obtain the Genocchi poly- nomials of higher order can be expressed by the Bernoulli polynomials of higher order when mis even

Corollary 2.5. Form∈N(m even), n∈N0, α∈C, we get G(α)n (mx) = (−2)αmn−α X

v1,v2,...,vm−1>0

α v1, v2, ..., vm−1

(−1)rBnα x+ r

m

. Also by applyingα= 1, in corollary 2.5 we obtain the following assertion that is one of the most remarkable identities in area of Genocchi polynomials.

Corollary 2.6. Form∈N, n∈N0, we obtain

Gn(mx) =−2mn−1

m−1

X

k=0

(−1)kBn

x+ k

m

m is even.

Obviously, the result of Corollary 2.6 is analogous with the well-known Raabe’s multiplication formula. Now, we present explicit evaluations of Zn(l)(m;λ), Zn(l)(λ),Zn(m) by Apostol-Genocchi polynomials.

(9)

Theorem 2.7. Form, n, l∈N0, λ∈C, we have

Zn(l)(m;λ) = 2−l

l

X

j=0

l j

(−1)j(m+1)λmj+l (n+ 1)l

n+l

X

k=0

n+l k

G(j)k (mj+l;λ)G(l−j)n+l−k(λ) where (n)0= 1,(n)k=n(n+ 1)...(n+k−1).

Proof. By definition ofZn(l)(m;λ), we calculate the following sum P

n=0Zn(l)(m;λ)tn!n =

X

n=0

h(−1)l X

06v1,v2,...,vm6l v1+v2+...+vm=l

l v1, v2, ..., vm

(−λ)λ1+2λ2+...+mλm(v1+2v2+...+mvm)nitn n!

= (−1)l X

06v1,v2,...,vm6l v1+v2+...+vm=l

l v1, v2, ..., vm

(−λet)λ1+2λ2+...+mλm

= (λet−λ2e2t+...+ (−1)m+1λmemt)l

=

(−1)m+1λm+1e(m+1)t

λet+ 1 + λet λet+ 1

l

= (2t)−l

l

X

j=0

l j

h2t(−1)m+1λm+1e(m+1)t λet+ 1

ijh 2tλet λet+ 1

il−j

= (2t)−l

l

X

j=0

l j

(−1)j(m+1)λmj+l

X

n=0

G(j)n (mj+l;λ)tn n!

X

n=0

G(l−j)n (λ)tn n!

= 2−l

X

n=0

l X

j=0

j l

)(−1)j(m+1)λmj+l (n+ 1)l

n+l

X

k=0

n+l k

G(j)k (mj+l;λ)G(l−j)n+l−k(λ) tn

n!

by comparing the coefficients of tn!n on both sides, proof will be complete.

As a direct result, usingλ= 1 in Theorem 2.7, we derive an explicit rep- resentation of multiple alternating sums Zn(l)(m), in terms of the Genocchi polynomials of higher order. We also deduce their special cases and applica- tions which lead to the corresponding results for the Genocchi polynomials.

Corollary 2.8. Form, n, l∈N0, the following formula holds true in terms of the Genocchi polynimials

Zn(l)(m) = 2−l

l

X

j=0

l j

(−1)j(m+1) (n+ 1)l

n+l

X

k=0

n+l k

G(j)k (mj+l)Gl−jn+l−k where (n)0= 1,(n)k=n(n+ 1)...(n+k−1).

(10)

Next we investigate some of the recursive formulas for the Apostol-Genocchi numbers of higher order that are analogous to the results of Howard [7, 8, 9]

and we deduce that they constitute a useful special case.

Theorem 2.9. Let m be odd, n, l∈N0 , λ∈C, then we have

mnG(l)nm)−mlG(l)n (λ) = (−1)l−1

n

X

k=0

n k

mkG(l)km)Zn−k(l) (m−1;λ).

Proof. By takingx= 0, α=lin (1), wherer =v1+ 2v2+...+ (m−1)vm−1

we obtain

mlG(l)n (λ) =mn X

v1,v2,...,vm−1>0

l v1, v2, ..., vm−1

(−λ)rG(l)n ( r m, λm) But we know

G(l)n (x;λ) =

n

X

k=0

n k

G(l)k (λ)xn−k So, we obtain

mlG(l)n (λ) = mn X

v1,v2,...,vm−1>0

l v1, v2, ..., vm−1

(−λ)r

n

X

k=0

n k

G(l)km) r

m n−k

=

n

X

k=0

n k

mkG(l)km) X

06v1,v2,...,vm−16l

l v1, v2, ..., vm−1

(−λ)rrn−k

=

n

X

k=0

n k

mkG(l)km) X

06v1,v2,...,vm−16l v1+v2+...vm−1=l

l v1, v2, ..., vm−1

(−λ)rrn−k+mnG(l)nm)

= (−1)l

n

X

k=0

n k

mkG(l)km)Zn−k(l) (m−1;λ) +mnG(l)nm)

So proof is complete.

Furthermore, we derive some well-known results (see [10]) involving Genoc- chi polynomials of higher order and Genocchi polynomials which we state here. By setting λ= 1,l= 1 in Theorem 2.9, we get Corollaries 2.10, 2.11, respectively.

Corollary 2.10. Let m be odd, n, l∈N0 , then we have

(mn−ml)G(l)n = (−1)l−1

n

X

k=0

n k

G(l)k Zn−k(l) (m−1).

Corollary 2.11. Let m be odd, n∈N0 , λ∈C, then we have

mnGnm)−mGn(λ) =

n

X

k=0

n k

mkGkm)Zn−k(m−1;λ).

(11)

Also by setting λ = 1 in Corollary 2.11, we get the following assertion that is analogous to the formula of Howard in terms of Genocchi numbers.

Corollary 2.12. For m be odd, n, l∈N0 , λ∈C, we obtain

(mn−m)Gn=

n

X

k=0

n k

mkGkZn−k(m−1) .

Next, we investigate the generalization of Howard’s formula in terms of Apostol-Genocchi numbers, whenm is even.

Theorem 2.13. Let m be even,n, l∈N0, λ∈C, the following formula

mlG(l)n (λ)−(−2)lmnBn(l)m) = 2l

n

X

k=0

n k

mkBk(l)m)Zn−k(l) (m−1;λ) holds true, where r=v1+ 2v2+...+ (m−1)vm−1.

Proof. We have

G(l)n (λ) = (−2)lmn−l X

v1,v2,...,vm−1>0

l v1, v2, ..., vm−1

(−λ)rBn(l)( r m, λm) But we know

B(l)n (x;λ) =

n

X

k=0

n k

Bk(l)(λ)xn−k So we get

mlG(l)n (λ) = (−2)lmn X

v1,v2,...,vm−1>0

l v1, v2, ..., vm−1

(−λ)r

n

X

k=0

n k

Bk(l)m)r m

n−k

= (−2)l

n

X

k=0

n k

mkBk(l)m) X

v1,v2,...,vm−1>0

l v1, v2, ..., vm−1

(−λ)rrn−k

= 2l

n

X

k=0

n k

mkBk(l)m)Zn−k(l) (m−1;λ) + (−2)lmnBn(l)m) So we obtain

mlG(l)n (λ)−(−2)lmnBn(l)m) = 2l

n

X

k=0

n k

mkBk(l)m)Zn−k(l) (m−1;λ)

So the proof is complete.

Also by lettingλ= 1 in Theorem 2.13, we obtain the following assertion.

Corollary 2.14. Let m be even,n, l∈N0,then we get

mlG(l)n −(−2)lmnBn(l) = 2l

n

X

k=0

n k

mkBn(l)Zn−k(l) (m−1)

(12)

Here we present a recurrence relation for Apostol-Genocchi numbers of higher order.

Theorem 2.15. Let n, k>1, then we have G(n+1)k (λ) = 2kG(n)k−1(λ)−

2− 2k n

G(n)k (λ) Proof. Let us put Gn(t;λ) =

2t λet+1

n

. Then Gn(t;λ) is the generat- ing function of higher order Apostol-Genocchi numbers. The derivative G0(t;λ) = dtdGn(t;λ) is equal to

n1

t − λet λet+ 1

Gn(t;λ) = n

tGn(t;λ)−nGn(t;λ) + n

λet+ 1Gn(t;λ) and

tG0n(t;λ) =nGn(t;λ)−ntGn(t;λ) +n

2Gn+1(t) so we obtain

G(n)k (λ)

(k−1)! =nG(n)k (λ)

k! −nG(n)k−1(λ) (k−1)! +n

2

G(n+1)k (λ) k!

fork>1. This formula can written as G(n+1)k (λ) = 2kG(n)k−1(λ)−

2− 2k n

G(n)k (λ)

so proof is complete.

3. generalized apostol genocchi polynomials with a, b, c parameters

In this section we investigate some recurrence formulas for generalized Apostol-Genocchi polynomials with a, b, c parameters . In 2003, Cheon [3] rederived several known properties and relations involving the classical Bernoulli polynomials Bn(x) and the classical Euler polynomials En(x) by making use of some standard techniques based upon series rearrangement as well as matrix representation. Srivastava and Pinter [36] followed Cheon’s work [3] and established two relations involving the generalized Bernoulli polynomials Bn(α)(x) and the generalized Euler polynomials En(α)(x). So, we will study further the relations between generalized Bernoulli polyno- mials with a, b parameters and Genocchi polynomials with the methods of generating function and series rearrangement.

Theorem 3.1. Let x ∈ R and n >0. For every positive real number a, b and c such thata6=b and b >0, we have

G(α)n (a, b;λ) =G(α)n

αlna lna−lnb;λ

(lnb−lna)n−α

(13)

Proof. We know 2t

λbt+at α

=

X

n=0

G(α)n (a, b;λ)tn n!

= 1

aαt

2t λet(lnb−lna)+ 1

α

= e−tαlna 2t(lnb−lna) λet(lnb−lna)+ 1

α

× 1

(lnb−lna)α

= 1

(lnb−lna)α

X

n=0

G(α)n αlna lna−lnb;λ

(lnb−lna)ntn n!

So by comparing the coefficients of tn!n on both sides, we get G(α)n (a, b;λ) =G(α)n

αlna lna−lnb;λ

(lnb−lna)n−α.

Theorem 3.2. Let x ∈ R and n >0. For every positive real number a, b and c such thata6=b and b >0, we have

G(α)n (x;a, b, c;λ) =G(α)n

−αlna+xlnc lnb−lna , λ

(lnb−lna)n−α Proof. We have

X

n=0

G(α)n (x;a, b, c;λ) =

2t λbt+at

α

cxt

= 1

αat

2t λet(lnb−lna)+ 1

α

cxt

= et(−αlna+xlnc) 2t λet(lnb−lna)+ 1

α

= 1

(lnb−lna)α

X

n=0

G(α)n −αlna+xlnc lnb−lna , λ

(lnb−lna)ntn n!.

So by comparing the coefficient of tn!n on both sides, we get G(α)n (x;a, b, c;λ) =G(α)n −αlna+xlnc

lnb−lna , λ

(lnb−lna)n−α

Therefore proof is complete.

The generalized Apostal-Genocchi polynomials of higher orderG(α)n (x;a, b, c;λ) possess a number of interesting properties which we state here.

Theorem 3.3. Let a, b, c∈R+ (a6=b) andx∈R, then (2) G(α)n (x+ 1;a, b, c;λ) =

n

X

k=0

n k

(lnc)n−kG(α)k (x;a, b, c;λ)

(14)

(3) G(α)n (x+α;a, b, c;λ) =G(α)n x;a

c,b c, c;λ

(4) G(α)n (α−x;a, b, c;λ) =G(α)n

−x;a c,b

c, c;λ

(5) G(α+β)n (x+y;a, b, c;λ) =

k

X

r=0

k r

G(α)k−r(x;a, b, c;λ)G(β)r (y;a, b, c;λ)

(6) ∂l

∂xl{G(α)n (x;a, b, c;λ)}= n!

(n−`)!(lnc)`G(α)n−`(x;a, b, c;λ) (7)

Z t

s

G(α)n (x;a, b, c;λ)dx= 1 (n+ 1) lnc

h

G(α)n+1(t;a, b, c;λ)−G(α)n+1(s;a, b, c;λ) i

Proof. We know

X

n=0

G(α)n (x+ 1;a, b, c;λ)tn

n! = t

λbt+at α

.c(x+1)t

=

X

n=0

X

k=0

G(α)k (x;a, b, c;λ)(lnc)ntn+k n!k!

=

X

n=0

X

k=0

G(α)k (x;a, b, c;λ)(lnc)n−k tn+k (n−k)!k!

So comparing the coefficients of tn on both sides, we arrive at the result (2) asserted by Theorem 3.3. Similary, by simple manipulations, leads us to the result (3), (4) and (5) of Theorem 3.3 and by successive differentiation with respect to x and then using the principle of mathematical induction on ` ∈ N0, we obtain the formula (6). Also, by taking ` = 1 in (6) and integrating both sides with respect tox, we get the formula (7).

Remark 3.4. Let a, b, c∈R+ (a6=−b) and x∈R, by differentiating both sides of the following generating function

X

n=0

Gαn(x;a, b, c;λ)tn

n! = tα (λetln(ab)+ 1)α

et(xlnc−xlna),

We get,

αλln(b a)

n

X

k=0

n k

(lnb)kG(α+1)n−k (x;a, b, c;λ) = (α−n)G(α)n (x;a, b, c;λ)

+ n(xlnc−αlna)G(α)n−1(x;a, b, c;λ).

(15)

Remark 3.5. GI-Sang Cheon and H. M. Srivastava in [3, 20] investigated the classical relationship between Bernoulli and Euler polynomials as follows

Bn(x) =

n

X

k=0 k6=1

n k

BkEn−k(x)

by applying a similar Srivastava’s method in [20] we obtain the following result for generalized Bernoulli polynomials and Genocchi numbers

Bn(x+y, a, b) = 1 2

n

X

k=0

1 n−k+ 1

n k

[Bk(y, a, b) +Bk(y+ 1, a, b)]Gn−k(x),

Gn(x+y) = 1 2

n

X

k=0

n k

[Gk(y) +Gk(y+ 1)]En−k(x), so, because we have

Gn(y+ 1) +Gn(y) = 2nyn−1, we obtain

Gn(x+y) =

n

X

k=0

k n

k

yk−1En−k(x) (y6= 0).

4. multiplication theorem for 2-variable Genocchi polynomial We apply the method of generating function, which are exploited to de- rive further classes of partial sums involving generalized many index many variable polynomials. In introduction we introduced 2-variable Genocchi polynomial. An application of 2-variable Genocchi polynomials is relevant to the multiplication theorems. In this section we develop the multiplication theorem for 2-variable Genocchi polynomial which yields a deeper insight into the effectiveness of this type of generalizations.

Theorem 4.1. Let x, y∈R+ and m be odd, we obtain

Gn(mx, py, λ) =mn−1

m−1

X

k=0

λk(−1)kHGn x+ k

m, py m2, λm Proof. We know

X

n=0

Gn(mx, py, λ)tn

n! = 2temxt+pyt2 λet+ 1

and handing the R. H. S of the above equations, we defined

X

n=0

Gn(mx, py, λ)tn

n! = 2temxt λmemt+ 1

λmemt+ 1 λet+ 1 epyt2

(16)

By noting that 2temxt λmemt+ 1

λmemt+ 1

λet+ 1 epyt2 =

m−1

X

k=0

1

m(−1)kλk

X

q=0

tqmq q! Gq

x+k

m, λmX

r=0

t2rpr r! yr We get

X

n=0

Gn(mx, py, λ)tn n! =

X

n=0

trmn−1

m−1

X

k=0

(−1)kλk

[n2]

X

r=0

Gn−2r(x+mk, λm) (n−2r)!r!

py m2

r

Also, by simple computation we realize that

HGn(x, y, λ) =

[n2]

X

s=0

ysGn−2s(x, λ) s!(n−2s)!

So, we obtain

Gn(mx, py, λ) =mn−1

m−1

X

k=0

(−1)kλkHGn x+ k

m, py m2, λm

Therefore proof is complete.

Also, by a similar method, we get the following remark.

Remark 4.2. Let m be odd and x, y∈R+, we get

HGn(mx, m2y, λ) =mn−1

m−1

X

`=0

(−1)`λ`HGn x+ `

m, y, λm .

Acknowledgments: The authors would like to thank the referees and editors and specially Maria Rohaly for their many valuable comments and suggestions.

References

[1] T. M. Apostol,On the Lerch Zeta function, Pacific. J. Math. No. 1, 1951, 161-167.

[2] I. N. Cangul, H. Ozden and Y. Simsek,A new approach to q-Genocchi numbers and their interpolation functions, Nonlinear Analysis: Theory, Methods and Applications, Vol.71, 2009, 793-799.

[3] G. S. Cheon,A note on the Bernoulli and Euler polynomials. Appl. Math. Lett. Vol.

16, No.3, 2003, 365-368.

[4] D. Dumont and G. Viennot,A Combinatorial Interpretation of the Seidel Generation of Genocchi Numbers, Annals of Discrete Mathematics, Vol.6, 1980, 77-87.

[5] B. N. Guo and F. Qi,Generalization of Bernoulli polynomials, J. Math. Ed. Sci. Tech.

33, No. 3, 2002, 428-431.

[6] S. Herrmann,Genocchi numbers and f-vectors of simplicial balls, European Journal of Combinatorics, Vol.29, Issue 5, 2008, 1087-1091.

[7] F. T. Howard,A sequence of numbers related to the exponential function, Duke. Math.

J.34, 1967, 599-616.

[8] F. T. Howard,Explicit formulas for degenerate Bernoulli numbers, Disc. Math, Vol.

162, Issue 1-3, 1996, 175-185.

(17)

[9] F. T. Howard, M. Cenkci,Notes on degenerate numbers, Disc. Math, Vol.307, Issues 19-20, 2007, 2359-2375.

[10] T. Kim, On the q-extension of Euler and Genocchi numbers, J. Math. Anal. Appl, Vol.326, Issue 2, 2007, 1458-1465.

[11] T. Kim and S.H. Rim ,Some q-Bernoulli numbers of higher order associated with the p-adic q-integrals. Indian J. Pure. Appl. Math.32, 2001, 1565-1570.

[12] G. D. Liu, H. M. Srivastava,Explicit formulas for the No¨urland polynomialsB(x)n and b(x)n , Comp. Math. Appl, Vol.51, Issue 9-10, 2006, 1377-1384.

[13] H. Liu and W. Wang, Some identities on the Bernoulli, Euler and Genocchi poly- nomials via power sums and alternate power sums, Discrete Mathematics, Vol.309, Issue 10, 2009, 3346-3363.

[14] S. D. Lin and H. M. Srivastava, Some families of the Hurwitz-Lerch Zeta function and associated fractional derivative and other integral repre-sentations. Appl. Math.

Comput,154, 2004, 725-733.

[15] Q. M. Luo,Some results for the q-Bernoulli and q-Euler polynomials, J. Math. Anal.

Apl, Vol.363, Issue 1, 2010, 7-18.

[16] Q. M. Luo,The multiplication formulas for the Apostol-Bernoulli and Apostol-Euler polynomials of higher order, Integral Transforms and Special Functions, Vol.20, Issue 5, 2009, 377-391.

[17] Q. M. Luo, B. N. Guo, F. Qi, and L. Debnath,Generalization of Bernoulli numbers and polynomials, IJMMS, Vol.2003, Issue 59, 2003, 3769-3776.

[18] Q. M. Luo, F. Qi, and L. Debnath,Generalizations of Euler numbers and polynomials, IJMMS. Vol. 2003, Issue 61, 2003(3893-3901)

[19] Q. M. Luo,q-Extensions for the Apostol-Genocchi Polynomials, General Mathematics Vol.17, No. 2 ,2009, 113-125.

[20] Q. M. Luo and H. M. Srivastava, Some relationships between the Apostol-Bernoulli and Apostol-Euler polynomials Computers and Mathematics with Applications, Vol.

51, Issues 3-4, 2006, 631-642.

[21] P. J. McCarthy , Some irreducibility theorems for Bernoulli polynomials of higher order, Duke Math. J. Vol.27, No. 3 ,1960, 313-318.

[22] J. Riordan and P. R. Stein, Proof of a conjecture on Genocchi numbers, Discrete Mathematics, Vol. 5, Issue 4, 1973, 381-388.

[23] S. H. Rim, K. H. Park and E. J. Moon, On Genocchi Numbers and Polynomials, Abstract and Applied Analysis. Vol.2008.

[24] B. Y. Rubinstein and L. G. Fel,Restricted partition functions as Bernoulli and Euler- ian polynomials of higher order, Ramanujan Journal, Vol.11, No. 3, 2006, 331-347.

[25] C. S. Ryoo,A numerical computation on the structure of the roots of q-extension of Genocchi polynomials, Applied Mathematics Letters, Vol.21, Issue 4, 2008, 348-354.

[26] C. S. Ryoo,A numerical computation on the structure of the roots of (h,q)-extension of Genocchi polynomials, Mathematical and Computer Modelling, Vol.49, Issues 3-4, 2009, 463-474.

[27] Y. Simsek,q-Hardy-Berndt type sums associated with q-Genocchi type zeta and q-l- functions, Nonlinear Analysis: Theory, Methods and Applications, Vol.71, Issue 12, 2009, 377-395.

[28] H. M. Srivastava,Some formulae for the q-Bernoulli and Euler polynomials of higher order, J. Math. Anal. Appl. Vol.273, Issue 1, 2002, 236-242.

[29] J. Zeng and J. Zhou,A q-analog of the Seidel generation of Genocchi numbers, Eu- ropean Journal of Combinatorics, Vol.27, Issue 3, 2006, 364-381.

[30] L. C. Jang, T. Kim, D. H. Lee, and D. W. Park,An application of polylogarithms in the analogue of Genocchi numbers, NNTDM, Vol.7, Issue 3, 2000, 66-70.

(18)

[31] Y. Simsek, I. N. Cangul, V. Kurt, and D. Kim, q-Genocchi numbers and polynomials associated with q-Genocchi-type l-functions, Adv. Difference Equ, doi:10.11555.2008/85750

[32] Jeff Remmel, Ascent Sequences, 2 + 2-free posets, Upper Triangular Matrices, and Genocchi numbers, Workshop on Combinatorics, Enumeration, and Invariant The- ory,George Mason University, Virginia, 2010.

[33] Anders Claesson, Sergey Kitaev, Kari Ragnarsson, Bridget Eileen Tenner, Boolean complexes for Ferrers graphs, arXiv:0808.2307v3

[34] Michael Domaratzki,Combinatorial Interpretations of a Generalization of the Genoc- chi Numbers, Journal of Integer Sequences, Vol.7, 2004.

[35] Qiu-Ming Luo, H.M. Srivastava,Some generalizations of the Apostol-Genocchi poly- nomials and the Stirling numbers of the second kind, Appl. Math. Comput. (2011), doi:10.1016/j.amc.2010.12.048.

[36] H.M. Srivastava and A. Pinter,Remarks on some relationships between the Bernoulli and Euler polynomials, Applied Math. Letter. Vol.17, 2004, 375-380.

[37] B. Kurt, The multiplication formulas for the Genocchi polynomials of higher order.

Proc. Jangjeon Math. Soc. Vol.13, No.1, 2010, 89-96.

[38] G. Dattoli, S. Lorenzutta and C. Cesarano,Bernoulli numbers and polynomials from a more general point of view. Rend. Mat. Appl. Vol.22, No.7, 2002, 193-202.

参照

関連したドキュメント

[2] Johann Cigler, Fibonacci polynomials, generalized Stirling numbers, and Bernoulli, Genocchi and tangent numbers, http://arxiv.org/abs/1103.2610.. [3] Johann Cigler,

For the authors, it is extremely surprising that some kind of discussion as in the previous paragraphs has not been included in the papers or books that define and study the

van Lint raised the problem whether the number 120 is the unique (positive) integer n for which the set { 1, 3, 8, 120 } constitutes a solution for Diophantus’ problem above; in

First of all, we reduce the problem (1.3)–(1.5) to an equivalent integral equa- tion by Green’s function of a boundary value problem for a second order ordinary differential

In this paper, using some classical inequalities, several inequalities involving zeros and coefficients of polynomials with real zeros have been obtained and the main result has

In the paper, the authors derive an explicit formula for derivative polynomials of the tangent function, deduce an explicit for- mula for tangent numbers, pose an open problem

Then the number of dormant torally indigenous bundles on a general curve of type (g, r ) over an algebraically closed field of characteristic p is given as the ( p − 2)nd value of

Chen first pointed out the relation between the exponent of convergence of distinct fixed points and the rate of growth of solutions of second order linear differential equations