Abstract
Eight basic identities of symmetry in three variables, which are related to degenerate Euler polynomials and alternating generalized falling factorial sums, are derived. These are the degenerate versions of the symmetric identities in three variables obtained in a previous paper. The derivations of identities are based on the p-adic integral expression of the generating function for the degenerate Euler polynomials and the quotient of integrals that can be expressed as the exponential generating function for the alternating generalized falling factorial sums. Those eight basic identities and most of their corollaries are new, since there have been results only about identities of symmetry in two variables.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction and Preliminaries
Let p be a fixed odd prime. Throughout this paper, \(\mathbb {Z}_{p},\mathbb {Q}_{p},\mathbb {C}_{p}\) will denote respectively the ring of p-adic integers, the field of p-adic rational numbers, and the completion of the algebraic closure of \(\mathbb {Q}_{p}\). For a continuous function \(f:\mathbb {Z}_{p}\rightarrow \mathbb {C}_{p}\), the p-adic fermionic integral of f is defined by Kim as
Then, it is easy to see that
where \(f_{1}(z)=f(z+1)\).
Let \(|\cdot |_{p}\) be the normalized absolute value of \(\mathbb {C}_{p}\) with \(|p|_{p}=\frac{1}{p}\). Throughout this paper, we assume that \(\lambda ,t\in \mathbb {C}_{p}\) satisfy
Then, as \(|\lambda t|_{p}<p^{-\frac{1}{p-1}}\), \(|\log (1+\lambda t)|_{p}=|\lambda t|_{p}\) and hence \(\left| \frac{1}{\lambda }\log \left( 1+\lambda t\right) \right| _{p}=\left| t\right| _{p}<p^{-\frac{1}{p-1}}\). Thus, \(f(t)=(1+\lambda t)^{\frac{z}{\lambda }}=e^{\frac{z}{\lambda }\log (1+\lambda t)}\) is a well-defined analytic function on \(\mathbb {Z}_{p}\). Applying (1) to this f, we obtain the p-adic integral representation of the generating function for the degenerate Euler numbers \(\mathcal {E}_{n}(\lambda )\):
Thus, we have the p-adic integral representation of the generating function for the degenerate Euler polynomials \(\mathcal {E}_{n}(\lambda ,x)\):
Note here that \(\mathcal {E}_{n} (\lambda )=\mathcal {E}_{n} (\lambda ,0)\),
where \(E_{n} (x)\) is the ordinary Euler polynomial.
The generalized falling factorial \((x\mid \lambda )_{n}\) is defined as \((x\mid \lambda )_{n}=x(x-\lambda )\cdots\) \((x-(n-1)\lambda )\), for \(n>0\) and \((x\mid \lambda )_{0}=1\). Let \(\tau _{k}(\lambda ,n)\) be the alternating generalized falling factorial sum given by
As special cases, we have
In addition
where \(T_{k}(n)\) denotes the alternating kth power sum of the first \(n+1\) nonnegative integers, namely
In addition, from (1) and (2), we get: for any odd positive integer w
Many authors have done much work on identities of symmetry in two variables involving Bernoulli polynomials or Euler polynomials or q-Bernoulli polynomials or q-Euler polynomials. The reader may refer to the papers (Bayad et al. 2011; Deeba and Rodriguez 1991; Howard 1995; Kim 2008, 2009b; Ozden et al. 2008a, b; Ozden and Simsek 2008; Simsek 2010; Tuenter 2001; Yang 2008). Especially, Kim (2008) is the first paper, where a p-adic approach is introduced. In connection with Bernoulli polynomials and power sums, and also with Euler polynomials and alternating power sums, these results were generalized in Kim (2011) and Kim (2009a) to obtain identities of symmetry involving three variables in contrast to the previous works involving just two variables. These have been done also for the q-Bernoulli and q-Euler polynomials (cf. Kim and Kim 2014; Kim et al. 2015) and for the higher order Bernoulli and higher order Euler polynomials (cf. Kim et al. 2013a, b; Kim 2011, 2012). It turns out that this extension gives not only new identities for three variables, but also those for two variables by specializing one of the variables as 1.
In this paper, we will produce eight basic identities of symmetry in three variables \(w_{1}, w_{2}, w_{3}\) related to degenerate Euler polynomials and alternating generalized falling factorial sums (cf. (48), (49), (52), (55), (59), (61), (63), (64)). These and most of their corollaries seem to be new, since there have been results only about identities of symmetry in two variables in the literature (cf. Kim and Kim 2015, 2016, 2017). These abundance of symmetries shed new light even on the existing identities in two variables. For instance, it has been known that (12) and (13) are equal and (14) and (15) are so (cf. [Kim and Kim 2015, Theorems 1, 2]). In fact, (12)–(15) are all equal, as they can be derived from one and the same p-adic integral. This was neglected to mention in Kim and Kim (2015). All of these were obtained as corollaries (cf. Cor. 4.9, 4.12, 4.15) to some of the basic identities by specializing the variable \(w_{3}\) as 1. Those would not be unearthed if more symmetries had not been available. The degenerate Euler polynomials were introduced in Carlitz (1979) and Wu and Pan (2014). In view of (4) and (7), all the symmetric identities in this paper approach to the corresponding ones in Kim (2009a), as \(\lambda\) tends to zero, and hence, our symmetric identities are the degenerate versions of the identities in Kim (2009a). Similar results about identities of symmetry in three variables for degenerate Bernoulli polynomials and generalized falling factorial sums were obtained in Kim et al. (2016).
Let \(w_{1},w_{2}\) be any odd positive integers. Then, we have
The derivations of identities are based on the p-adic integral expression of the generating function for the degenerate Euler polynomials in (3) and the quotient of integrals in (10) that can be expressed as the exponential generating function for the alternating generalized falling factorial sums.
2 Several Types of Quotients of Fermionic Integrals
Here, we will introduce several types of quotients of p-adic fermionic integrals on \(\mathbb Z_{p}\) or \(\mathbb Z_{p}^{3}\) from which some interesting identities follow owing to the built-in symmetries in \(w_{1},w_{2},w_{3}\). In the following, \(w_{1},w_{2},w_{3}\) are all positive integers and all of the explicit expressions of integrals in (21), (23), (25), and (27) are obtained from the identity in (2). In below, \(\mathrm{d}\mu _{-1}(x_{1}, x_{2}, x_{3})\) denotes \(\mathrm{d}\mu _{-1}(x_{1})\mathrm{d}\mu _{-1}(x_{2})\mathrm{d}\mu _{-1}(x_{3})\).
-
(a)
Type \(\Lambda _{23}^{i}\) (for \(i=0,1,2,3\))
$$\begin{aligned} I(\Lambda _{23}^{i})&=\frac{\int _{\mathbb Z_{p}^{3}}(1+\lambda t)^{\frac{1}{\lambda }(w_{2}w_{3}x_{1}+w_{1}w_{3}x_{2}+w_{1}w_{2}x_{3} +w_{1}w_{2}w_{3}(\sum _{j=1}^{3-i}y_{j}))} \mathrm{d}\mu _{-1}(x_{1}, x_{2}, x_{3})}{(\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}w_{3}x_{4}}\mathrm{d}\mu _{-1}(x_{4}))^{i}}\end{aligned}$$(19)$$\begin{aligned}&=\frac{2^{3-i}(1+\lambda t)^{\frac{w_{1}w_{2}w_{3}}{\lambda }(\sum _{j=1}^{3-i}y_{j})}((1+\lambda t)^{\frac{w_{1}w_{2}w_{3}}{\lambda }}+1)^{i}}{((1+\lambda t)^{\frac{w_{2}w_{3}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{1}w_{3}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{1}w_{2}}{\lambda }}+1)}; \end{aligned}$$(20) -
(b)
Type \(\Lambda _{13}^{i}\) (for \(i=0,1,2,3\))
$$\begin{aligned} I(\Lambda _{13}^{i})&=\frac{\int _{\mathbb Z_{p}^{3}}(1+\lambda t)^{\frac{1}{\lambda }(w_{1}x_{1}+w_{2}x_{2}+w_{3}x_{3}+w_{1}w_{2}w_{3} (\sum _{j=1}^{3-i}y_{j}))}\mathrm{d}\mu _{-1}(x_{1})\mathrm{d}\mu _{-1}(x_{2})\mathrm{d}\mu _{-1}(x_{3})}{(\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}w_{3}x_{4}}\mathrm{d}\mu _{-1}(x_{4}))^{i}}\end{aligned}$$(21)$$\begin{aligned}&=\frac{2^{3-i}(1+\lambda t)^{\frac{w_{1}w_{2}w_{3}}{\lambda }(\sum _{j=1}^{3-i}y_{j})}((1+\lambda t)^{\frac{w_{1}w_{2}w_{3}}{\lambda }}+1)^{i}}{((1+\lambda t)^{\frac{w_{1}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{2}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{3}}{\lambda }}+1)}; \end{aligned}$$(22) -
(c-0)
Type \(\Lambda _{12}^{0}\)
$$\begin{aligned} I(\Lambda _{12}^{0})&=\int _{\mathbb Z_{p}^{3}}(1+\lambda t)^{\frac{1}{\lambda }(w_{1}x_{1}+w_{2}x_{2}+w_{3}x_{3} +w_{2}w_{3}y+w_{1}w_{3}y+w_{1}w_{2}y)} \mathrm{d}\mu _{-1}(x_{1}, x_{2}, x_{3}) \end{aligned}$$(23)$$\begin{aligned}&=\frac{8(1+\lambda t)^{\frac{y}{\lambda }(w_{2}w_{3}+w_{1}w_{3}+w_{1}w_{2})}}{((1+\lambda t)^{\frac{w_{1}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{2}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{3}}{\lambda }}+1)}; \end{aligned}$$(24) -
(c-1)
Type \(\Lambda _{12}^{1}\)
$$\begin{aligned} I(\Lambda _{12}^{1})&=\frac{\int _{\mathbb Z_{p}^{3}}(1+\lambda t)^{\frac{1}{\lambda }(w_{1}x_{1}+w_{2}x_{2} +w_{3}x_{3})}\mathrm{d}\mu _{-1}(x_{1})\mathrm{d}\mu _{-1}(x_{2})\mathrm{d}\mu _{-1}(x_{3})}{\int _{\mathbb Z_{p}^{3}}(1+\lambda t)^{\frac{1}{\lambda }(w_{2}w_{3}z_{1}+w_{1}w_{3}z_{2} +w_{1}w_{2}z_{3})}\mathrm{d}\mu _{-1}(z_{1})\mathrm{d}\mu _{-1}(z_{2})\mathrm{d}\mu _{-1}(z_{3})} \end{aligned}$$(25)$$\begin{aligned}&=\frac{((1+\lambda t)^{\frac{w_{2}w_{3}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{1}w_{3}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{1}w_{2}}{\lambda }}+1)}{((1+\lambda t)^{\frac{w_{1}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{2}}{\lambda }}+1)((1+\lambda t)^{\frac{w_{3}}{\lambda }}+1)}. \end{aligned}$$(26)
All of the above p-adic integrals of various types are invariant under all permutations of \(w_{1},w_{2},w_{3}\) as one can see either from p-adic integral representations in (20), (22), (24), and (26) or from their explicit evaluations in (21), (23), (25), and (27).
3 Identities for Degenerate Euler Polynomials
In the following, \(w_{1},w_{2},w_{3}\) are all odd positive integers except for (a-0) and (c-0), where they are any positive integers.
First, let us consider Type \(\Lambda _{23}^{i}\), for each \(i=0,1,2,3.\) The following results can be easily obtained from (3), (10), and (11).
-
(a-0)
$$\begin{aligned} I(\Lambda _{23}^{0})&=\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}(x_{1}+w_{1}y_{1})}\mathrm{d}\mu _{-1}(x_{1})\end{aligned}$$(27)$$\begin{aligned}&\quad \times \int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{3}(x_{2}+w_{2}y_{2})}\mathrm{d}\mu _{-1}(x_{2})\nonumber \\&\quad \times \int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}(x_{3}+w_{3}y_{3})}\mathrm{d}\mu _{-1}(x_{3})\nonumber \\&=\left( \sum _{k=0}^{\infty }\frac{\mathcal {E}_{k}\left( \frac{\lambda }{w_{2}w_{3}}, w_{1}y_{1}\right) }{k!}\left( w_{2}w_3{}t\right) ^k\right) \left( \sum _{l=0}^{\infty }\frac{\mathcal {E}_{l}\left( \frac{\lambda }{w_{1}w_{3}}, w_{2}y_{2}\right) }{l!}(w_{1}w_{3}t)^l\right) \nonumber \\&\quad \times \left( \sum _{m=0}^{\infty }\frac{\mathcal {E}_{m}(\frac{\lambda }{w_{1}w_{2}},w_{3}y_{3})}{m!}(w_{1}w_{2}t)^m\right) \nonumber \\&=\sum _{n=0}^{\infty } \bigg (\sum _{k+l+m=n}\left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) \mathcal {E}_{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}y_{1}\bigg )\mathcal {E}_{l}\bigg (\frac{\lambda }{w_{1}w_{3}}, w_{2}y_{2}\bigg )\nonumber \\&\quad \times \mathcal {E}_{m}\bigg (\frac{\lambda }{w_{1}w_{2}},w_{3}y_{3}\bigg ) w_{1}^{l+m}w_{2}^{k+m}w_{3}^{k+l}\bigg )\frac{t^{n}}{n!}, \end{aligned}$$(28)
where the inner sum is over all nonnegative integers k, l, m, with \(k+l+m=n\), and
$$\begin{aligned} \left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) =\frac{n!}{k!l!m!}. \end{aligned}$$ -
(a-1)
Here, we write \(I(\Lambda _{23}^{1})\) in two different ways:
-
(1)
$$\begin{aligned} I(\Lambda _{23}^{1})&=\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}(x_{1}+w_{1}y_{1})}\mathrm{d}\mu _{-1}(x_{1})\end{aligned}$$(29)$$\begin{aligned}&\times \int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{3}(x_{2}+w_{2}y_{2})} \mathrm{d}\mu _{-1}(x_{2})\frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}x_{3}}\mathrm{d}\mu _{-1}(x_{3})}{\int _{Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}w_{3}x_{4}}\mathrm{d}\mu _{-1}(x_{4})}\nonumber \\&=\bigg (\sum _{k=0}^{\infty }{\mathcal {E}_{k}\bigg (\frac{\lambda }{w_{2}w_{3}}, w_{1}y_{1}\bigg )\frac{(w_{2}w_{3}t)^k}{k!}}\bigg ) \bigg (\sum _{l=0}^{\infty }{\mathcal {E}_{l}\bigg (\frac{\lambda }{w_{1}w_{3}}, w_{2}y_{2}\bigg )\frac{(w_{1}w_{3}t)^l}{l!}}\bigg )\nonumber \\&\times \bigg (\sum _{m=0}^{\infty }{\tau _{m}\bigg (\frac{\lambda }{w_{1}w_{2}},w_{3}-1\bigg ) \frac{(w_{1}w_{2}t)^m}{m!}}\bigg )\nonumber \\&=\sum _{n=0}^{\infty }\bigg (\sum _{k+l+m=n}\left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) \mathcal {E}_{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}y_{1}\bigg )\mathcal {E}_{l}\bigg (\frac{\lambda }{w_{1} w_{3}},w_{2}y_{2}\bigg )\nonumber \\&\times \tau _{m}\bigg (\frac{\lambda }{w_{1}w_{2}},w_{3}-1\bigg )w_{1}^{l+m} w_{2}^{k+m}w_{3}^{k+l})\frac{t^{n}}{n!}. \end{aligned}$$(30)
-
(2)
Invoking (10), (30) can also be written as
$$\begin{aligned}&I(\Lambda _{23}^{1})=\sum _{i=0}^{w_{3}-1}(-1)^i\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}(x_{1}+w_{1}y_{1})} \mathrm{d}\mu _{-1}(x_{1})\nonumber \\&\times \int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{3}(x_{2}+w_{2}y_{2}+\frac{w_{2}}{w_{3}}i)}\mathrm{d}\mu _{-1}(x_{2})\nonumber \\&=\sum _{i=0}^{w_{3}-1}(-1)^i\bigg (\sum _{k=0}^{\infty }\mathcal {E}_{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}y_{1}\bigg )\frac{(w_{2}w_{3}t)^k}{k!}\bigg ) \bigg (\sum _{l=0}^{\infty }\mathcal {E}_{l}\bigg (\frac{\lambda }{w_{1}w_{3}}, w_{2}y_{2}+\frac{w_{2}}{w_{3}}i\bigg )\frac{(w_{1}w_{3}t)^l}{l!}\bigg ) \nonumber \\&=\sum _{n=0}^{\infty }\bigg (w_{3}^{n}\sum _{k=0}^{n}\left( {\begin{array}{c}n\\ k\end{array}}\right) \mathcal {E}_{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}y_{1}\bigg )\sum _{i=0}^{w_{3}-1} (-1)^i\mathcal {E}_{n-k}\bigg (\frac{\lambda }{w_{1}w_{3}},w_{2}y_{2} +\frac{w_{2}}{w_{3}}i\bigg )w_{1}^{n-k}w_{2}^{k}\bigg )\frac{t^{n}}{n!}. \end{aligned}$$(31)
-
(1)
-
(a-2)
Here, we write \(I(\Lambda _{23}^{2})\) in three different ways:
-
(1)
$$\begin{aligned} I(\Lambda _{23}^{2})&=\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}(x_{1}+w_{1}y_{1})}\mathrm{d}\mu _{-1}(x_{1})\end{aligned}$$(32)$$\begin{aligned}&\quad \times \frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{3}x_{2}}\mathrm{d}\mu _{-1}(x_{2})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}w_{3}x_{4}}\mathrm{d}\mu _{-1}(x_{4})}\frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}x_{3}}\mathrm{d}\mu _{-1}(x_{3})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}w_{3}x_{4}}\mathrm{d}\mu _{-1}(x_{4})} \nonumber \\&=\bigg (\sum _{k=0}^{\infty }{\mathcal {E}_{k}\bigg (\frac{\lambda }{w_{2}w_{3}}, w_{1}y_{1}\bigg )\frac{(w_{2}w_{3}t)^k}{k!}}\bigg ) \bigg (\sum _{l=0}^{\infty }{\tau _{l}\bigg (\frac{\lambda }{w_{1}w_{3}},w_{2}-1\bigg ) \frac{(w_{1}w_{3}t)^l}{l!}}\bigg ) \nonumber \\&\times \bigg (\sum _{m=0}^{\infty }{\tau _{m}\bigg (\frac{\lambda }{w_{1}w_{2}}, w_{3}-1\bigg )\frac{(w_{1}w_{2}t)^m}{m!}}\bigg )\nonumber \\&=\sum _{n=0}^{\infty }\bigg (\sum _{k+l+m=n}\left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) \mathcal {E}_{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}y_{1}\bigg )\tau _{l} \bigg (\frac{\lambda }{w_{1}w_{3}},w_{2}-1\bigg )\nonumber \\&\times \tau _{m}\bigg (\frac{\lambda }{w_{1}w_{2}},w_{3}-1\bigg )w_{1}^{l+m}w_{2}^{k+m}w_{3}^{k+l}\bigg ) \frac{t^{n}}{n!}. \end{aligned}$$(33)
-
(2)
Invoking (10), (33) can also be written as
$$\begin{aligned} I(\Lambda _{23}^{2})&=\sum _{i=0}^{w_{2}-1}(-1)^i\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}(x_{1}+w_{1}y_{1} +\frac{w_{1}}{w_{2}}i)}\mathrm{d}\mu _{-1}(x_{1})\end{aligned}$$(34)$$\begin{aligned}&\quad \times \frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}x_{3}}\mathrm{d}\mu _{-1}(x_{3})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}w_{3}x_{4}}\mathrm{d}\mu _{-1}(x_{4})}\nonumber \\&=\sum _{i=0}^{w_{2}-1}(-1)^i\bigg (\sum _{k=0}^{\infty }\mathcal {E}_{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}y_{1}+\frac{w_{1}}{w_{2}}i\bigg ) \frac{(w_{2}w_{3}t)^k}{k!}\bigg )\nonumber \\&\times \bigg (\sum _{l=0}^{\infty }\tau _{l}\bigg (\frac{\lambda }{w_{1}w_{2}}, w_{3}-1\bigg )\frac{(w_{1}w_{2}t)^l}{l!}\bigg )\nonumber \\&=\sum _{n=0}^{\infty }\bigg (w_{2}^{n}\sum _{k=0}^{n}\left( {\begin{array}{c}n\\ k\end{array}}\right) \sum _{i=0}^{w_{2}-1}(-1)^i\mathcal {E}_{k}\bigg (\frac{\lambda }{w_{2}w_{3}}, w_{1}y_{1}+\frac{w_{1}}{w_{2}}i\bigg )\nonumber \\&\times \tau _{n-k}\bigg (\frac{\lambda }{w_{1}w_{2}},w_{3}-1\bigg )w_{1}^{n-k}w_{3}^{k}\bigg ) \frac{t^{n}}{n!}. \end{aligned}$$(35) -
(3)
Invoking (10) once again, (35) can be written as
$$\begin{aligned}&I(\Lambda _{23}^{2})\nonumber \\&=\sum _{i=0}^{w_{2}-1}\sum _{j=0}^{w_{3}-1}(-1)^{i+j}\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}(x_{1}+w_{1}y_{1} +\frac{w_{1}}{w_{2}}i+\frac{w_{1}}{w_{3}}j)}\mathrm{d}\mu _{-1}(x_{1})\nonumber \\&=\sum _{i=0}^{w_{2}-1}\sum _{j=0}^{w_{3}-1}(-1)^{i+j} \sum _{n=0}^{\infty }\mathcal {E}_{n}\bigg (\frac{\lambda }{w_{2}w_{3}}, w_{1}y_{1}+\frac{w_{1}}{w_{2}}i+\frac{w_{1}}{w_{3}}j\bigg ) \frac{(w_{2}w_{3}t)^n}{n!}\nonumber \\&=\sum _{n=0}^{\infty }\bigg ((w_{2}w_{3})^n\sum _{i=0}^{w_{2}-1} \sum _{j=0}^{w_{3}-1}(-1)^{i+j}\mathcal {E}_{n}\bigg (\frac{\lambda }{w_{2}w_{3}}, w_{1}y_{1}+\frac{w_{1}}{w_{2}}i+\frac{w_{1}}{w_{3}}j\bigg )\bigg )\frac{t^{n}}{n!}. \end{aligned}$$(36)
-
(1)
-
(a-3)
$$\begin{aligned} I(\Lambda _{23}^{3})&=\frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{w_{2}w_{3}x_{1}t}\mathrm{d}\mu _{-1}(x_{1})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{w_{1}w_{2}w_{3}x_{4}t}\mathrm{d}\mu _{-1}(x_{4})} \frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{w_{1}w_{3}x_{2}t}\mathrm{d}\mu _{-1}(x_{2})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{w_{1}w_{2}w_{3}x_{4}t}\mathrm{d}\mu _{-1}(x_{4})}\nonumber \\&\quad \times \frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{w_{1}w_{2}x_{3}t}\mathrm{d}\mu _{-1}(x_{3})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{w_{1}w_{2}w_{3}x_{4}t}\mathrm{d}\mu _{-1}(x_{4})}\nonumber \\&=\bigg (\sum _{k=0}^{\infty }\tau _{k}\bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}-1\bigg ) \frac{(w_{2}w_{3}t)^{k}}{k!}\bigg ) \bigg (\sum _{l=0}^{\infty }\tau _{l}\bigg (\frac{\lambda }{w_{1}w_{3}},w_{2}-1\bigg ) \frac{(w_{1}w_{3}t)^{l}}{l!}\bigg )\nonumber \\&\quad \times \bigg (\sum _{m=0}^{\infty }\tau _{m}\bigg (\frac{\lambda }{w_{1}w_{2}}, w_{3}-1\bigg )\frac{(w_{1}w_{2}t)^{m}}{m!}\bigg )\nonumber \\&=\sum _{n=0}^{\infty }\sum _{k+l+m=n}^{}\bigg (\left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) \tau _{k} \bigg (\frac{\lambda }{w_{2}w_{3}},w_{1}-1\bigg )\tau _{l}\bigg (\frac{\lambda }{w_{1}w_{3}}, w_{2}-1\bigg )\nonumber \\&\quad \times \tau _{m}\bigg (\frac{\lambda }{w_{1}w_{2}}, w_{3}-1\bigg )w_{1}^{l+m}w_{2}^{k+m}w_{3}^{k+l}\bigg )\frac{t^{n}}{n!}. \end{aligned}$$(37)
-
(b)
For Type \(\Lambda _{13}^{i}~(i=0,1,2,3)\), we may consider the analogous things to the ones in (a-0), (a-1), (a-2), and (a-3). However, these do not lead us to new identities. Indeed, if we substitute \(w_{2}w_{3},w_{1}w_{3},w_{1}w_{2}\), respectively, for \(w_{1},w_{2},w_{3}\) in (20), this amounts to replacing \(\lambda\) by \(\frac{\lambda }{w_{1}w_{2}w_{3}}\) and t by \(w_{1}w_{2}w_{3}t\) in (22). Therefore, upon replacing \(w_{1},w_{2},w_{3}\), respectively ,by \(w_{2}w_{3},w_{1}w_{3},w_{1}w_{2}\), and then replacing \(\lambda\) by \(w_{1}w_{2}w_{3}\lambda\) and dividing by \((w_{1}w_{2}w_{3})^n\), in each of the expressions of Theorem 4.1 through Corollary 4.15, we will get the corresponding symmetric identities for Type \(\Lambda _{13}^{i}~(i=0,1,2,3)\).
-
(c-0)
$$\begin{aligned} I(\Lambda _{12}^{0})&=\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}(x_{1}+w_{2}y)}\mathrm{d}\mu _{-1}(x_{1}) \int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}(x_{2}+w_{3}y)}\mathrm{d}\mu _{-1}(x_{2})\nonumber \\&\quad \times \int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{3}(x_{3}+w_{1}y)}\mathrm{d}\mu _{-1}(x_{3})\nonumber \\&=\bigg (\sum _{n=0}^{\infty }\frac{\mathcal {E}_{k}(\frac{\lambda }{w_{1}}, w_{2}y)}{k!}(w_{1}t)^{k}\bigg )\bigg (\sum _{l=0}^{\infty }\frac{\mathcal {E}_{l} (\frac{\lambda }{w_{2}},w_{3}y)}{l!}(w_{2}t)^{l}\bigg )\bigg (\sum _{m=0}^{\infty } \frac{\mathcal {E}_{m}(\frac{\lambda }{w_{3}},w_{1}y)}{m!}(w_{3}t)^{m}\bigg )\nonumber \\&=\sum _{n=0}^{\infty }\bigg (\sum _{k+l+m=n}\left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) \mathcal {E}_{k} \bigg (\frac{\lambda }{w_{1 }},w_{2}y\bigg )\mathcal {E}_{l}\bigg (\frac{\lambda }{w_{2 }},w_{3}y\bigg )\mathcal {E}_{m}\bigg (\frac{\lambda }{w_{3 }},w_{1}y\bigg )w_{1}^{k}w_{2}^{l}w_{3}^{m}\bigg )\frac{t^n}{n!}. \end{aligned}$$(38)
-
(c-1)
$$\begin{aligned} I(\Lambda ^{1}_{12})&=\frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}x_{1}}\mathrm{d}\mu _{-1}(x_{1})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{1}w_{2}z_{3}}\mathrm{d}\mu _{-1}(z_{3})} \times \frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}x_{2}}\mathrm{d}\mu _{-1}(x_{2})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{2}w_{3}z_{1}}\mathrm{d}\mu _{-1}(z_{1})}\nonumber \\&\quad \times \frac{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{3}x_{3}}\mathrm{d}\mu _{-1}(x_{3})}{\int _{\mathbb Z_{p}}(1+\lambda t)^{\frac{1}{\lambda }w_{3}w_{1}z_{2}}\mathrm{d}\mu _{-1}(z_{2})}\nonumber \\&=\bigg (\sum _{k=0}^{\infty }\tau _{k}\bigg (\frac{\lambda }{w_{1 }},w_{2}-1\bigg )\frac{(w_{1}t)^{k}}{k!}\bigg ) \bigg (\sum _{l=0}^{\infty }\tau _{l}\bigg (\frac{\lambda }{w_{2 }},w_{3}-1\bigg )\frac{(w_{2}t)^{l}}{l!}\bigg )\nonumber \\&\quad \times \bigg (\sum _{m=0}^{\infty }\tau _{m}\bigg (\frac{\lambda }{w_{3 }},w_{1}-1\bigg )\frac{(w_{3}t)^{m}}{m!}\bigg )\nonumber \\&=\sum _{n=0}^{\infty }\bigg (\sum _{k+l+m=n}^{}\left( {\begin{array}{c}n\\ k,l,m\end{array}}\right) \tau _{k} \bigg (\frac{\lambda }{w_{1}},w_{2}-1\bigg )\tau _{l} \bigg (\frac{\lambda }{w_{2}},w_{3}-1\bigg )\nonumber \\&\quad \times \tau _{m}\bigg (\frac{\lambda }{w_{3}}, w_{1}-1\bigg )w_{1}^{k}w_{2}^{l}w_{3}^{m}\bigg ) \frac{t^{n}}{n!}. \end{aligned}$$(39)
4 Main Theorems
As we noted earlier in the last paragraph of Sect. 2, the various types of quotients of p-adic fermionic integrals are invariant under any permutation of \(w_{1},w_{2},w_{3}\). Therefore, the corresponding expressions in Sect. 3 are also invariant under any permutation of \(w_{1},w_{2},w_{3}\). Thus, our results about identities of symmetry will be immediate consequences of this observation.
However, not all permutations of an expression in the above yield distinct ones. In fact, as these expressions are obtained by permuting \(w_{1}, w_{2},w_{3}\) in a single equation labeled by them, there is a natural transitive action of the group \(S_{3}\) on those set of expressions, and hence, they are in bijective correspondence with a quotient of \(S_{3}\). In particular, the number of possible distinct expressions are 1, 2, 3, or 6. Indeed, (a-0), (a-1(1)), (a-1(2)), and (a-2(2)) give the full six identities of symmetry, (a-2(1)) and (a-2(3)) yield three identities of symmetry, and (c-0) and (c-1) give two identities of symmetry, while the expression in (a-3) yields no identities of symmetry.
Here, we will just consider the cases of Theorems 4.8 and 4.17, leaving the others as easy exercises for the reader. As for the case of Theorem 4.8, in addition to (54)–(56), we get the following three ones:
However, by interchanging l and m, we see that (41), (42), and (43) are, respectively, equal to (54), (55), and (56).
As to Theorem 4.17, in addition to (64) and (65), we have
However, (44) and (45) are equal to (64), as we can see by applying the permutations \(k\rightarrow l,l\rightarrow m,m\rightarrow k\) for (44) and \(k\rightarrow m,l\rightarrow k,m\rightarrow l\) for (45). Similarly, we see that (46) and (47) are equal to (65), by applying permutations \(k\rightarrow l,l\rightarrow m,m\rightarrow k\) for (46) and \(k\rightarrow m,l\rightarrow k,m\rightarrow l\) for (47).
Theorems 4.1, 4.2, 4.5, 4.8, 4.11, 4.14, 4.16, and 4.17 follow respectively from the considerations in (a-0), (a-1(1)), (a-1(2)), (a-2(1)), (a-2(2)), (a-2(3)), (c-0), and (c-1) in Sect. 3.
Theorem 4.1
Let \(w_{1},w_{2},w_{3}\) be any positive integers. Then, the following expression is invariant under any permutation of \(w_{1},w_{2},w_{3}\), so that it gives us six symmetries:
Theorem 4.2
Let \(w_{1},w_{2},w_{3}\) be any odd positive integers. Then, the following expression is invariant under any permutation of \(w_{1},w_{2},w_{3}\), so that it gives us six symmetries:
Putting \(w_{3}=1\) in (49), we get the following corollary.
Corollary 4.3
Let \(w_{1},w_{2}\) be any odd positive integers:
Letting further \(w_{2}=1\) in (50), we have the following corollary.
Corollary 4.4
Let \(w_{1}\) be any odd positive integer:
Theorem 4.5
Let \(w_{1},w_{2},w_{3}\) be any odd positive integers. Then, the following expression is invariant under any permutation of \(w_{1},w_{2},w_{3}\), so that it gives us six symmetries:
Letting \(w_{3}=1\) in (52), we obtain alternative expressions for the identities in (50).
Corollary 4.6
Let \(w_{1},w_{2}\) be any odd positive integers:
Putting further \(w_{2}=1\) in (53), we have the alternative expressions for the identities for (51).
Corollary 4.7
Let \(w_{1}\) be any odd positive integer:
Theorem 4.8
Let \(w_{1},w_{2},w_{3}\) be any odd positive integers. Then, we have the following three symmetries in \(w_{1},w_{2},w_{3}\):
Putting \(w_{3}=1\) in (54)–(56), we get the following corollary.
Corollary 4.9
Let \(w_{1},w_{2}\) be any odd positive integers:
Letting further \(w_{2}=1\) in (57), we get the following corollary.
Corollary 4.10
Let \(w_{1}\) be any odd positive integer:
Theorem 4.11
Let \(w_{1},w_{2},w_{3}\) be any odd positive integers. Then, the following expression is invariant under any permutation of \(w_{1},w_{2},w_{3}\), so that it gives us six symmetries:
Putting \(w_{3}=1\) in (59), we obtain the following corollary. In Sect. 1, the identities in (57), (60), and (62) are combined to give those in (12)–(19):
Corollary 4.12
Let \(w_{1},w_{2}\) be any odd positive integers:
Letting further \(w_{2}=1\) in (60), we get the following corollary. This is the multiplication formula for Euler polynomials (cf. [Kim and Kim 2015, Page 6]) together with the new identity mentioned in (58).
Corollary 4.13
Let \(w_{1}\) be any odd positive integer:
Theorem 4.14
Let \(w_{1},w_{2},w_{3}\) be any odd positive integers. Then, we have the following three symmetries in \(w_{1},w_{2},w_{3}\):
Letting \(w_{3}=1\) in (61), we have the following corollary.
Corollary 4.15
Let \(w_{1},w_{2}\) be any odd positive integers:
Theorem 4.16
Let \(w_{1},w_{2},w_{3}\) be any positive integers. Then, we have the following two symmetries in \(w_{1},w_{2},w_{3}\):
Theorem 4.17
Let \(w_{1},w_{2},w_{3}\) be any odd positive integers. Then, we have the following two symmetries in \(w_{1},w_{2},w_{3}\):
Putting \(w_{3}=1\) in (64) and (65), we get the following corollary.
Corollary 4.18
Let \(w_{1},w_{2}\) be any odd positive integers:
References
Bayad A, Kim T, Choi J, Kim YH, Lee B (2011) On the symmetry properties of the generalized higher-order Euler polynomials. J Appl Math Inf 29(1–2):511–516
Carlitz L (1979) Degenerate Stirling, Bernoulli and Eulerian numbers. Util Math 15:51–88
Deeba E, Rodriguez D (1991) Stirling’s and Bernoulli numbers. Am Math Mon 98:423–426
Howard FT (1995) Applications of a recurrence for the Bernoulli numbers. J Number Theory 52:157–172
Kim T (2008) Symmetry \(p\)-adic invariant integral on \(\mathbb{Z}_{p}\) for Bernoulli and Euler polynomials. J Differ Equ Appl 14:1267–1277
Kim T (2009a) Symmetry of power sum polynomials and multivariate fermionic \(p\)-adic invariant integral on \({\mathbb{Z}}_p\). Russ J Math Phys 16:93–96
Kim T (2009b) Symmetry identities for the twisted generalized Euler polynomials. Adv Stud Cotemp Math (Kyungshang) 19:151–155
Kim T (2011) An identity of the symmetry for the Frobenius–Euler polynomials associated with the fermionic \(p\)-adic invariant \(q\)-integrals on \({\mathbb{Z}}_p\). Rocky Mt J Math 41:239–247
Kim DS, Kim T (2014) Three variable symmetric identities involving Carlitz-type q-Euler polynomials. Math Sci 8:147–152
Kim DS, Kim T (2015) Identities of symmetry for the generalized degenerate Euler polynomials. In: Singh VK, Srivastava HM, Venturino E, Resch M, Gupta V (eds) Modern Mathematical Methods and High Performance Computing in Science and Technology, M3HPCST. Springer Proceedings in Mathematics & Statistics, vol 171. Springer, Ghaziabad, India, p 33–43. doi:10.1007/978-981-10-1454-3
Kim DS, Kim T (2016) Symmetric identities of higher-order degenerate q-Euler polynomials. J Nonlinear Sci Appl 9:443–451
Kim DS, KIM T (2017) Symmetric identities of higher-order degenerate Euler polynomials. arXiv:1704.04025
Kim DS, Lee N, Na J, Park KH (2011) Identities of symmetry for higher-order Euler polynomials in three variables (II). J Math Anal Appl 379:388–400
Kim DS, Lee N, Na J, Park KH (2012) Identities of symmetry for higher-order Euler polynomials in three variables (I). Adv Stud Contemp Math (Kyungshang) 22(1):51–74
Kim DS, Lee N, Na J, Park KH (2013a) Abundant symmetry for higher-order Bernoulli polynomials (I). Adv Stud Contemp Math (Kyungshang) 23(3):461–482
Kim DS, Lee N, Na J, Park KH (2013b) Abundant symmetry for higher-order Bernoulli polynomials (II). Proc Jangjeon Math Soc 16(3):359–378
Kim T, Dolgy DV, Kim DS (2015) Some identities of \(q\)-Bernoulli polynomials under symmetry group \(S_3\). J Nonlinear Convex Anal 9:1869–1880
Kim T, Kim DS, Dolgy DV, Kwon HI, Seo JJ (2016) Identities of symmetry for degenerate Bernoulli polynomials and generalized falling factorial sums. Global J Pure Appl Math 12(5):4363–4383
Ozden H, Simsek Y (2008) A new extension of \(q\)-Euler numbers and polynomials related to their interpolation functions. Appl Math Lett 21(9):934–939
Ozden H, Cangul IN, Simsek Y (2008a) Multivariate interpolation functions of higher-order \(q\)-Euler numbers and their applications. Abstr Appl Anal Article ID 390857
Ozden H, Cangul IN, Simsek Y (2008b) Remarks on sum of products of \((h,q)\)-twisted Euler polynomials and numbers. J Inequal Appl Article ID 816129
Simsek Y (2010) Complete sum of products of \((h, q)\)-extension of Euler polynomials and numbers. J Differ Equ Appl 16(11):1331–1348
Tuenter H (2001) A symmetry of power sum polynomials and Bernoulli numbers. Am Math Mon 108:258–261
Wu M, Pan H (2014) Sums of products of the degenerate Euler numbers. Adv Differ Equ 2014:40
Yang S (2008) An identity of symmetry for the Bernoulli polynomials. Discrete Math 308:550–554
Acknowledgements
The first author is appointed as a chair professor at Tianjin Polytechnic University by Tianjin City in China from August 2015 to August 2019. The authors would like to express their sincere gratitude to referees for their valuable comments and information.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kim, T., Kim, D.S. Identities of Symmetry for Degenerate Euler Polynomials and Alternating Generalized Falling Factorial Sums. Iran J Sci Technol Trans Sci 41, 939–949 (2017). https://doi.org/10.1007/s40995-017-0326-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40995-017-0326-6
Keywords
- Degenerate Euler polynomial
- Alternating generalized falling factorial sum
- Fermionic p-adic integral
- Identities of symmetry