Open AccessArticle

General Linear Recurrence Sequences and Their Convolution Formulas

Paolo Emilio Ricci

and

Pierpaolo Natalini

^2,*

Section of Mathematics, International Telematic University UniNettuno, Corso Vittorio Emanuele II, 39, 00186 Roma, Italy

Dipartimento di Matematica e Fisica, Università degli Studi Roma Tre, Largo San Leonardo Murialdo, 1, 00146 Roma, Italy

Author to whom correspondence should be addressed.

Axioms 2019, 8(4), 132; https://doi.org/10.3390/axioms8040132

Submission received: 29 September 2019 / Revised: 11 November 2019 / Accepted: 15 November 2019 / Published: 19 November 2019

(This article belongs to the Special Issue Differential and Difference Equations: A Themed Issue Dedicated to Prof. Hari M. Srivastava on the Occasion of his 80th Birthday)

Download Versions Notes

Abstract

We extend a technique recently introduced by Chen Zhuoyu and Qi Lan in order to find convolution formulas for second order linear recurrence polynomials generated by

{(\frac{1}{1 + a t + b t^{2}})}^{x} .

The case of generating functions containing parameters, even in the numerator is considered. Convolution formulas and general recurrence relations are derived. Many illustrative examples and a straightforward extension to the case of matrix polynomials are shown.

Keywords:

liner recursions; convolution formulas; Gegenbauer polynomials; Humbert polynomials; classical polynomials in several variables; classical number sequences

AMS 2010 Mathematics Subject Classifications:

33C99; 65Q30; 11B37

1. Introduction

Generating functions [1] constitute a bridge between continuous analysis and discrete mathematics. Linear recurrence relations are satisfied by many special polynomials of classical analysis. A wide scenario including special sequences of polynomials and numbers, combinatorial analysis, and application of mathematics is related to the above mentioned topics.

It would be impossible to list in the Reference section all of even the most important articles dedicated to these subjects. As a first example, we recall the Chebyshev polynomials of the first and second kind, which are powerful tools used in both theoretical and applied mathematics. Their links with the Lucas and Fibonacci polynomials have been studied and many properties have been derived. Connections with Bernoulli polynomials have been highlighted in [2].

In particular, the important calculation of sums of several types of polynomials have been recently studied (see e.g., [3,4,5] and the references therein). This kind of subject has attracted many scholars. For example, W. Zhang [6] proved an identity involving Chebyshev polynomials and their derivatives.

Fibonacci and Lucas polynomias and their extensions have been studied for a long time, in particular within the Fibonacci Association, which has contributed to the study of this and similar subjects. As an applications of a results proved by Y. Zhang and Z. Chen [3], Y. Ma and W. Zhang [4] obtained some identities involving Fibonacci numbers and Lucas numbers.

Convolution techniques are connected with combinatorial identities, and many results have been obtained in this direction [2,7,8]. Convolution sums using second kind Chebyshev polynomials are contained in [7].

Recently, Taekyun Kim et al. [8] studied properties of Fibonacci numbers by introducing the so called convolved Fibonacci numbers. By using the genereting function:

\begin{matrix} \begin{matrix} {(\frac{1}{1 - t - b t^{2}})}^{x} = \sum_{n = 0}^{\infty} p_{n} (x) \frac{t^{n}}{n!}, \end{matrix} \end{matrix}

for

x \in R

and

r \in N

, they proved the interesting relation

\begin{matrix} \begin{matrix} p_{n} (x) = \sum_{ℓ = 0}^{n} p_{ℓ} (r) p_{n - ℓ} (x - r) = \sum_{ℓ = 0}^{n} p_{n - ℓ} (r) p_{ℓ} (x - r) . \end{matrix} \end{matrix}

Furthermore, they derived a link between

p_{n} (x)

and a particular combination of sums of Fibonacci numbers, so that complex sums of Fibonacci numbers have been converted to the easier calculation of

p_{n} (x)

In a recent article Chen Zhuoyu and Qi Lan [9] introduced convolution formulas for second order linear recurrence sequences related to the generating function [1] of the type

f (t) = \frac{1}{1 + a t + b t^{2}},

deriving coefficient expressions for the series expansion of the function

f^{x} (t)

(x \in R)

. In this article, motivated by this research, we continue the study of possible applications of the considered method, by analyzing the general situation of a generating function of the type

G (t, x) = {(\frac{1}{1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}})}^{x},

and we deduce the recurrence relation for the generated polynomials.

Several illustrative examples are shown in Section 6. In the last section the results are extended, in a straightforward way, to the case of matrix polynomials.

2. Generating Functions

We start from the generating function considered by Chen Zhuoyu and Qi Lan:

\begin{matrix} \begin{matrix} G (t, x) = {(\frac{1}{1 + a t + b t^{2}})}^{x} = {[\frac{1}{(1 - α t) (1 - β t)}]}^{x} = \\ = \exp \{- x \log [(1 - α t) (1 - β t)]\}, \end{matrix} \end{matrix}

(1)

with

\begin{matrix} \begin{matrix} a = - (α + β), b = α β \end{matrix} \end{matrix}

(2)

\begin{matrix} \begin{matrix} G (t, x) = \sum_{k = 0}^{\infty} g_{k} (x; α, β) \frac{t^{k}}{k!} = \\ = \exp \{- x \log (1 - α t)\} \cdot \exp \{- x \log (1 - β t)\} = G_{α} (t, x) \cdot G_{β} (t, x), \end{matrix} \end{matrix}

(3)

where

\begin{matrix} G_{α} (t, x) = \exp [- x \log (1 - α t)] = \sum_{k = 0}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!}, \end{matrix}

(4a)

\begin{matrix} G_{β} (t, x) = \exp [- x \log (1 - β t)] = \sum_{k = 0}^{\infty} q_{k} (x, β) \frac{t^{k}}{k!} . \end{matrix}

(4b)

Note that, by Equation (2) we could write, in equivalent form:

\begin{matrix} \begin{matrix} g_{k} (x; α, β) = g_{k} (x; a, b), p_{k} (x, α) = p_{k} (x, a), q_{k} (x, β) = q_{k} (x, b), \end{matrix} \end{matrix}

(5)

but, in what follows, we put for shortness:

\begin{matrix} \begin{matrix} g_{k} (x; α, β) = g_{k} (x), p_{k} (x, α) = p_{k} (x), q_{k} (x, β) = q_{k} (x) . \end{matrix} \end{matrix}

(6)

By Equations (3), (4a) and (4b) we find the convolution formula:

\begin{matrix} \begin{matrix} g_{k} (x) = \sum_{h = 0}^{k} (\binom{k}{h}) p_{k - h} (x) q_{h} (x) . \end{matrix} \end{matrix}

(7)

3. Recurrence Relation

Note that

\begin{matrix} \begin{matrix} \frac{\partial G (t, x)}{\partial t} = \frac{\partial G_{α} (t, x)}{\partial t} \cdot G_{β} (t, x) + G_{α} (t, x) \cdot \frac{\partial G_{β} (t, x)}{\partial t} = \\ = (\frac{α x}{1 - α t} + \frac{β x}{1 - β t}) G (t, x) = - x (\frac{a + 2 b t}{1 + a t + b t^{2}}) G (t, x), \end{matrix} \end{matrix}

(8)

as can be derived directly from Equation (1).

Then we have

\begin{matrix} \begin{matrix} (1 + a t + b t^{2}) \frac{\partial G (t, x)}{\partial t} = - x a G (t, x) - 2 b x t G (t, x), \end{matrix} \end{matrix}

(9)

\begin{matrix} \begin{matrix} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k}}{k!} + a \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k + 1}}{k!} + b \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k + 2}}{k!} = \\ = - x a \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k}}{k!} - 2 b x \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k + 1}}{k!}, \end{matrix} \end{matrix}

that is

\begin{matrix} \begin{matrix} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k}}{k!} + a \sum_{k = 1}^{\infty} k g_{k} (x) \frac{t^{k}}{k!} + b \sum_{k = 2}^{\infty} k (k - 1) g_{k - 1} (x) \frac{t^{k}}{k!} = \\ = - a x \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k}}{k!} - 2 b x \sum_{k = 1}^{\infty} k g_{k - 1} (x) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

and therefore, we can conclude with the theorem:

Theorem 1.

The sequence

{g_{k} (x)}_{k \in N}

satisfies the linear recurrence relation

\begin{matrix} \begin{matrix} g_{k} (x) + a (x + k - 1) g_{k - 1} (x) + b (k - 1) (k + 2 x - 2) g_{k - 2} (x) = 0 . \end{matrix} \end{matrix}

(10)

3.1. Properties of the Basic Generating Function

We consider now a few properties of the basic generating functions

G_{α} (t, x)

. According to the definition (4a), the polynomials

p_{k} (x)

are recognized as associated Sheffer polynomials [10] and quasi-monomials, according to the Dattoli [11,12] definition.

3.1.1. Differential Equation

We have:

\begin{matrix} \begin{matrix} G_{α} (t, x) = \exp [- x H (t)] = \sum_{k = 0}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

(11)

where

\begin{matrix} \begin{matrix} H (t) = - \log (1 - α t), H^{'} (t) = \frac{α}{1 - α t}, \end{matrix} \end{matrix}

(12)

and its functional inverse is given by

\begin{matrix} \begin{matrix} H^{- 1} (t) = \frac{1}{α} (1 - e^{- t}), \end{matrix} \end{matrix}

(13)

so that, recalling the results by Y. Ben Cheikh [13], we find the derivative and multiplication operators of the quasi-monomials

p_{k} (x)

, in the form:

\begin{matrix} \begin{matrix} \hat{P} = \frac{1}{α} (1 - e^{- D_{x}}) \hat{M} = x H^{'} (H^{- 1} (D_{x})) = α x e^{D_{x}}, \end{matrix} \end{matrix}

(14)

and we can conclude that

Theorem 2.

The polynomials

p_{k} (x)

satisfy the differential equation:

\begin{matrix} \begin{matrix} \hat{M} \hat{P} p_{n} (x) = x (e^{D_{x}} - 1) p_{n} (x) = n p_{n} (x), \end{matrix} \end{matrix}

(15)

that is,

\forall n \geq 1

\begin{matrix} \begin{matrix} x (\frac{1}{n!} p_{n}^{(n)} + \frac{1}{(n - 1)!} p_{n}^{(n - 1)} + \dots + p_{n}^{'} (x)) = n p_{n} (x) . \end{matrix} \end{matrix}

(16)

3.1.2. Differential Identity

Differentiating Equation (11) with respect to x, we find

\begin{matrix} \begin{matrix} \frac{\partial G_{α} (t, x)}{\partial x} = - G_{α} (t, x) \log (1 - α t) = \sum_{k = 1}^{\infty} p_{k}^{'} (x, α) \frac{t^{k}}{k!} \end{matrix} \end{matrix}

(17)

that is

\begin{matrix} \begin{matrix} \sum_{k = 1}^{\infty} p_{k}^{'} (x, α) \frac{t^{k}}{k!} = - \log (1 - α t) \sum_{k = 0}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

\begin{matrix} \begin{matrix} \sum_{k = 1}^{\infty} p_{k}^{'} (x, α) \frac{t^{k}}{k!} = \sum_{k = 1}^{\infty} \frac{{(α t)}^{k}}{k} \sum_{k = 0}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!} = \sum_{k = 1}^{\infty} \frac{{(α t)}^{k}}{k} [1 + \sum_{k = 1}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!}], \end{matrix} \end{matrix}

\begin{matrix} \begin{matrix} \sum_{k = 1}^{\infty} p_{k}^{'} (x, α) \frac{t^{k}}{k!} = \sum_{k = 1}^{\infty} (k - 1)! α^{k} \frac{t^{k}}{k} + \sum_{k = 1}^{\infty} (k - 1)! α^{k} \frac{t^{k}}{k} \sum_{k = 1}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

\begin{matrix} \begin{matrix} \sum_{k = 1}^{\infty} p_{k}^{'} (x, α) \frac{t^{k}}{k!} = \sum_{k = 1}^{\infty} (k - 1)! α^{k} \frac{t^{k}}{k!} + \sum_{k = 1}^{\infty} (k - 1)! α^{k} \frac{t^{k}}{k!} \sum_{k = 1}^{\infty} p_{k} (x, α) \frac{t^{k}}{k!} = \\ = \sum_{k = 1}^{\infty} (k - 1)! α^{k} \frac{t^{k}}{k!} + \sum_{k = 1}^{\infty} \sum_{h = 1}^{k} (k - h - 1)! α^{k - h} p_{h} (x, α) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

so that we can conclude with the theorem:

Theorem 3.

The polynomials

p_{k} (x)

satisfy the differential identity:

\begin{matrix} \begin{matrix} p_{k}^{'} (x, α) = (k - 1)! α^{k} + \sum_{h = 1}^{k} (k - h - 1)! α^{k - h} p_{h} (x, α) . \end{matrix} \end{matrix}

(18)

3.2. Extension by Convolution

We now consider the case of a generating function of the type:

\begin{matrix} \begin{matrix} G (t, x) = {(\frac{1 + c t}{1 + a t + b t^{2}})}^{x} = \sum_{k = 0}^{\infty} q_{k} (x; c; a, b) = \frac{\sum_{k = 0}^{\infty} p_{k} (x; c)}{\sum_{k = 0}^{\infty} g_{k} (x; a, b)} . \end{matrix} \end{matrix}

(19)

A straightforward consequence is the convolution formula for the resulting polynomials:

\begin{matrix} p_{k} (x; c) = \sum_{h = 0}^{k} (\binom{k}{h}) g_{k - h} (x; a, b) q_{h} (x; c; a, b), \end{matrix}

(20)

so that the

q_{h} (x; c; a, b)

can be found recursively by solving the infinite system

\begin{matrix} \{\begin{matrix} q_{0} (x; c; a, b) = 1, \\ q_{k} (x; c; a, b) = p_{k} (x; a, b) - \sum_{h = 0}^{k - 1} (\binom{k}{h}) g_{k - h} (x; a, b) q_{h} (x; c; a, b) . \end{matrix} \end{matrix}

(21)

Noting that

p_{0} (x; a, b) = g_{0} (x; c; a, b) = 1

, the very first polynomials are given by

\begin{matrix} \begin{matrix} q_{0} (x; c; a, b) = 1, \\ q_{1} (x; c; a, b) = p_{1} (x; a, b) - g_{1} (x; c; a, b), \\ q_{2} (x; c; a, b) = p_{2} (x; a, b) - 2 g_{1} (x; c; a, b) p_{1} (x; a, b) + 2 g_{1}^{2} (x; c; a, b) - g_{2} (x; c; a, b), \\ q_{3} (x; c; a, b) = p_{3} (x; a, b) - 3 g_{1} (x; c; a, b) p_{2} (x; a, b) + 6 g_{1}^{2} (x; c; a, b) p_{1} (x; a, b) \\ - 6 g_{1}^{3} (x; c; a, b) + 6 g_{1} (x; c; a, b) g_{2} (x; c; a, b) - 3 g_{2} (x; c; a, b) p_{1} (x; a, b) \\ - g_{3} (x; c; a, b) . \end{matrix} \end{matrix}

(22)

Further values can be obtained by using symbolic computation.

4. The General Case

Note that the above results can be extended to the general case, considering the generating function:

\begin{matrix} \begin{matrix} G (t, x) = {(\frac{1}{1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}})}^{x} = {[\frac{1}{(1 - α_{1} t) (1 - α_{2} t) \dots (1 - α_{r} t)}]}^{x} = \\ = \exp \{- x \log [(1 - α_{1} t) (1 - α_{2} t) \dots (1 - α_{r} t)]\} = \sum_{k = 0}^{\infty} g_{k} (x; α_{1}, α_{2}, \dots, α_{r}) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

(23)

where

\begin{matrix} \begin{matrix} a_{1} = σ_{1} = - (α_{1} + α_{2} + \dots + α_{r}), \\ \dots, \\ a_{s} = σ_{r} = {(- 1)}^{s} \sum_{j_{1}, j_{2}, \dots, j_{s}} α_{j_{1}} α_{j_{2}} \dots α_{j_{s}}, \\ \dots, \\ a_{r} = σ_{r} = α_{1} α_{2} \dots α_{r}, \end{matrix} \end{matrix}

(24)

are the elementary symmetric functions of the zeros.

Putting as before:

\begin{matrix} \begin{matrix} G_{α_{h}} (t, x) = \exp [- x \log (1 - α_{h} t)] = \sum_{k = 0}^{\infty} p_{1, k} (x, α_{h}) \frac{t^{k}}{k!}, (h = 1, 2, \dots, r), \end{matrix} \end{matrix}

(25)

since

\begin{matrix} G (t, x) = G_{α_{1}} (t, x) \cdot G_{α_{2}} (t, x) \dots G_{α_{r}} (t, x), \end{matrix}

we find the result:

Theorem 4.

The sequence

{g_{k} (x)}_{k \in N}

satisfies the convolution formula:

\begin{matrix} \begin{matrix} g_{k} (x) = \sum_{\begin{matrix} k_{1} + k_{2} + \dots + k_{r} = k \\ 0 \leq k_{i} \leq k \end{matrix}} (\binom{k}{k_{1}, k_{2}, \dots, k_{r}}) p_{1, k_{1}} (x) p_{2, k_{2}} (x) \dots p_{r, k_{r}} (x), \end{matrix} \end{matrix}

(26)

where, according to our position,

\begin{matrix} \begin{matrix} g_{k} (x) = g_{k} (x; α_{1}, α_{2}, \dots, α_{r}), p_{1, k_{1}} (x) = p_{1, k_{1}} (x, α_{1}), \dots, p_{r, k_{r}} (x) = p_{r, k_{r}} (x, α_{r}) . \end{matrix} \end{matrix}

5. The General Recurrence Relation

From Equation (17) we find:

\begin{matrix} \begin{matrix} \frac{\partial G (t, x)}{\partial t} = - x (\frac{a_{1} + 2 a_{2} t + \dots + r a_{r} t^{r - 1}}{1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}}) G (t, x), \end{matrix} \end{matrix}

(27)

\begin{matrix} \begin{matrix} (1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}) \frac{\partial G (t, x)}{\partial t} = - x (a_{1} + 2 a_{2} t + \dots + r a_{r} t^{r - 1}) G (t, x), \end{matrix} \end{matrix}

\begin{matrix} \begin{matrix} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k}}{k!} + a_{1} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k + 1}}{k!} + a_{2} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k + 2}}{k!} + \dots + a_{r} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k + r}}{k!} = \\ = - a_{1} x \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k}}{k!} - 2 a_{2} x \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k + 1}}{k!} - \dots - r a_{r} x \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k + r - 1}}{k!}, \end{matrix} \end{matrix}

that is

\begin{matrix} \begin{matrix} \sum_{k = 0}^{\infty} g_{k + 1} (x) \frac{t^{k}}{k!} + a_{1} \sum_{k = 1}^{\infty} k g_{k} (x) \frac{t^{k}}{k!} + a_{2} \sum_{k = 2}^{\infty} k (k - 1) g_{k - 1} (x) \frac{t^{k}}{k!} + \dots \\ + a_{r} \sum_{k = r}^{\infty} k (k - 1) \dots (k - r + 1) g_{k - r + 1} (x) \frac{t^{k}}{k!} = \\ = - a_{1} x \sum_{k = 0}^{\infty} g_{k} (x) \frac{t^{k}}{k!} - 2 a_{2} x \sum_{k = 0}^{\infty} k g_{k - 1} (x) \frac{t^{k}}{k!} - \dots \\ - r a_{r} x \sum_{k = 0}^{\infty} k (k - 1) \dots (k - r + 2) g_{k - r + 1} (x) \frac{t^{k}}{k!} . \end{matrix} \end{matrix}

Therefore, we can conclude that

Theorem 5.

The sequence

{g_{k} (x)}_{k \in N}

satisfies the linear recurrence relation

\begin{matrix} \begin{matrix} g_{k} (x) + a_{1} (x + k - 1) g_{k - 1} (x) + a_{2} (k - 1) (2 x + k - 2) g_{k - 2} (x) + \dots \\ + a_{r} (k - 1) (k - 2) \dots (k - r + 1) (r x + k - r) g_{k - r} (x) = 0 . \end{matrix} \end{matrix}

(28)

Extension to the General Case

We now generalize the convolution formula in Section 3.2, putting for shortness

{[c]}_{r - 1} = c_{1}, c_{2}, \dots, c_{r - 1}, {[a]}_{r} = a_{1}, a_{2}, \dots, a_{r},

and considering the generating function:

\begin{matrix} \begin{matrix} G (t, x) = {(\frac{1 + c_{1} t + c_{2} t^{2} + \dots + c_{r - 1} t^{r - 1}}{1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}})}^{x} = \sum_{k = 0}^{\infty} q_{k} (x; {[c]}_{r - 1}; {[a]}_{r}) \frac{t^{k}}{k!} = \\ = \frac{\sum_{k = 0}^{\infty} p_{k} (x; {[c]}_{r - 1}) \frac{t^{k}}{k!}}{\sum_{k = 0}^{\infty} g_{k} (x; {[a]}_{r}) \frac{t^{k}}{k!}}, \end{matrix} \end{matrix}

(29)

so that we find the convolution formula:

\begin{matrix} p_{k} (x; {[c]}_{r - 1}) = \sum_{h = 0}^{k} (\binom{k}{h}) g_{k - h} (x; {[a]}_{r}) q_{h} (x; {[c]}_{r - 1}; {[a]}_{r}), \end{matrix}

(30)

and the

q_{h} (x; {[c]}_{r - 1}; {[a]}_{r})

can be found recursively by solving the infinite system

\begin{matrix} \{\begin{matrix} q_{0} (x; {[c]}_{r - 1}; {[a]}_{r}) = 1, \\ q_{k} (x; {[c]}_{r - 1}; {[a]}_{r}) = p_{k} (x; {[a]}_{r}) - \sum_{h = 0}^{k - 1} (\binom{k}{h}) g_{k - h} (x; {[a]}_{r}) q_{h} (x; {[c]}_{r - 1}; {[a]}_{r}) . \end{matrix} \end{matrix}

(31)

6. Illustrative Examples—Second Order Recurrences

Gegenbauer polynomials [14], defined by

$\begin{matrix} \begin{matrix} {(1 - 2 y t + t^{2})}^{- λ} = \sum_{k = 0}^{\infty} C_{k}^{(λ)} (y) t^{k}, \end{matrix} \end{matrix}$

$x = λ$ , $a = - 2 y$ , $b = 1$ , $g_{k} (λ; - 2 y, 1) = k! C_{k}^{(λ)} (y)$ .
Sinha polynomials [15], defined by

$\begin{matrix} \begin{matrix} {[1 - 2 y t + (2 y - 1) t^{2}]}^{- ν} = \sum_{k = 0}^{\infty} S_{k}^{(ν)} (y) t^{k}, \end{matrix} \end{matrix}$

$x = ν$ , $a = - 2 y$ , $b = (2 y - 1)$ , $g_{k} (ν; - 2 y, 2 y - 1) = k! S_{k}^{(ν)} (y)$ .
Fibonacci polynomials [16], defined by

$\begin{matrix} \begin{matrix} \frac{t}{1 - y t - t^{2}} = \sum_{k = 0}^{\infty} F_{k} (y) t^{k}, F_{k} (1) = F_{k} (Fibonacci numbers) . \end{matrix} \end{matrix}$

We have:

$\begin{matrix} \begin{matrix} \frac{t}{1 - y t - t^{2}} = t \sum_{k = 0}^{\infty} g_{k} (1; - y, - 1) \frac{t^{k}}{k!} = \sum_{k = 0}^{\infty} k! F_{k} (y) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}$

so that

$\begin{matrix} \begin{matrix} \sum_{k = 1}^{\infty} k g_{k - 1} (1; - y, - 1) \frac{t^{k}}{k!} = \sum_{k = 0}^{\infty} k! F_{k} (y) \frac{t^{k}}{k!} . \end{matrix} \end{matrix}$

Since $F_{0} (y) = 0$ , we find

$\begin{matrix} \begin{matrix} F_{k} (y) = \frac{1}{(k - 1)!} g_{k - 1} (1; - y, - 1) . \end{matrix} \end{matrix}$
Lucas polynomials [16], defined by

$\begin{matrix} \begin{matrix} \frac{2 - y t}{1 - y t - t^{2}} = \sum_{k = 0}^{\infty} L_{k} (y) t^{k}, L_{k} (1) = L_{k} (Lucas numbers) . \end{matrix} \end{matrix}$

We have:

$\begin{matrix} \begin{matrix} \frac{2 - y t}{1 - y t - t^{2}} = 2 \sum_{k = 0}^{\infty} g_{k} (1; - y, - 1) \frac{t^{k}}{k!} - y \sum_{k = 1}^{\infty} k g_{k - 1} (1; - y, - 1) \frac{t^{k}}{k!} = \sum_{k = 0}^{\infty} k! L_{k} (y) \frac{t^{k}}{k!} . \end{matrix} \end{matrix}$

Since $L_{0} (y) = 0$ , we find

$\begin{matrix} \begin{matrix} L_{k} (y) = (\frac{2}{k!} - \frac{y}{(k - 1)!}) g_{k - 1} (1, - y, - 1) . \end{matrix} \end{matrix}$

Illustrative Examples—Higher Order Recurrences

Humbert polynomials [14], defined by

$\begin{matrix} \begin{matrix} {(1 - 3 y t + t^{3})}^{- λ} = \sum_{k = 1}^{\infty} u_{k} (y) t^{k}, \end{matrix} \end{matrix}$

$x = λ$ , $a_{1} = - 3 y$ , $a_{2} = 0$ , $a_{3} = 1$ , $g_{k} (λ; - 3 y, 0, 1) = k! u_{k} (y)$ .
First kind Chebyshev polynomials in several variables [17,18,19,20], defined by

$\begin{matrix} \begin{matrix} \frac{r - (r - 1) u_{1} t + (r - 2) u_{2} t^{2} + \dots + {(- 1)}^{r - 1} u_{r - 1} t^{r - 1}}{1 - u_{1} t + u_{2} t^{2} - \dots + {(- 1)}^{r - 1} u_{r - 1} t^{r - 1} + {(- 1)}^{r} t^{r}} = \sum_{k = 0}^{\infty} T_{k} (u_{1}, \dots, u_{r - 1}) t^{k}, \end{matrix} \end{matrix}$

$\begin{matrix} x = 1, c_{1} = - \frac{r - 1}{r} u_{1}, \dots, c_{r - 1} = \frac{{(- 1)}^{r - 1}}{r} u_{r - 1}, a_{1} = - u_{1}, \dots, a_{r - 1} = {(- 1)}^{r - 1} u_{r - 1}, a_{r} = {(- 1)}^{r}, \\ q_{k} (1; {[c]}_{r - 1}; {[a]}_{r}) = \frac{1}{r} k! T_{k} (u_{1}, \dots, u_{r - 1}) . \end{matrix}$
Second kind Chebyshev polynomials in several variables [17,18,19,20], defined by

$\begin{matrix} \begin{matrix} \frac{1}{1 - u_{1} t + u_{2} t^{2} - \dots + {(- 1)}^{r - 1} u_{r - 1} t^{r - 1} + {(- 1)}^{r} t^{r}} = \sum_{k = 0}^{\infty} U_{k} (u_{1}, \dots, u_{r - 1}) t^{k}, \end{matrix} \end{matrix}$

$x = 1, a_{1} = - u_{1}, \dots, a_{r - 1} = {(- 1)}^{r - 1} u_{r - 1}, a_{r} = {(- 1)}^{r}, g_{k} (1; {[a]}_{r}) = k! U_{k} (u_{1}, \dots, u_{r - 1}) .$
Tribonacci polynomials [21], defined by

$\begin{matrix} \begin{matrix} \frac{t}{1 - y^{2} t - y t^{2} - t^{3}} = \sum_{k = 0}^{\infty} τ_{k} (y) t^{k} . \end{matrix} \end{matrix}$

We have:

$\begin{matrix} \begin{matrix} \frac{t}{1 - y^{2} t - y t^{2} - t^{3}} = t \sum_{k = 0}^{\infty} g_{k} (1, - y^{2}, - y, - 1) \frac{t^{k}}{k!} = \sum_{k = 0}^{\infty} k! τ_{k} (y) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}$

so that

$\begin{matrix} \begin{matrix} \sum_{k = 1}^{\infty} k g_{k - 1} (1, - y^{2}, - y, - 1) \frac{t^{k}}{k!} = \sum_{k = 0}^{\infty} k! τ_{k} (y) \frac{t^{k}}{k!} . \end{matrix} \end{matrix}$

Since $τ_{0} (y) = 0$ , we find

$\begin{matrix} \begin{matrix} τ_{k} (y) = \frac{1}{(k - 1)!} g_{k - 1} (1, - y^{2}, - y, - 1) . \end{matrix} \end{matrix}$

7. Extension to Matrix Polynomials

Extensions to Matrix polynomials have become a fashionable subject recently (see e.g., [22] and the references therein).

The above results can be easily extended to Matrix polynomials assuming, in Equations (1), (7), (10), (17), (20), and (22), instead of x, a complex

N \times N

matrix A, satisfying the condition:

A is stable, that is, denoting by

σ (A)

the spectrum of A, this results in:

\forall λ \in σ (A), ℜ λ > 0

Since all powers of a matrix A commute, even every matrix polynomial commute. More generally, if

σ (A) \subset Ω

, where

Ω

is an open set of the complex plane, for any holomorphic functions f and g, this results in:

\begin{matrix} f (A) g (A) = g (A) f (A), \end{matrix}

that is, the involved matrix functions commute.

Under these conditions, considering the generating function:

\begin{matrix} \begin{matrix} G (t, A) = {(\frac{1}{1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}})}^{A} = \\ = \exp \{- A \log [1 + a_{1} t + a_{2} t^{2} + \dots + a_{r} t^{r}]\} = \sum_{k = 0}^{\infty} g_{k} (A; a_{1}, a_{2}, \dots, a_{r}) \frac{t^{k}}{k!}, \end{matrix} \end{matrix}

(32)

recalling positions (18), and putting as before:

\begin{matrix} \begin{matrix} G_{α_{h}} (t, A) = \exp [- A \log (1 - α_{h} t)] = \sum_{k = 0}^{\infty} p_{1, k} (A, α_{h}) \frac{t^{k}}{k!}, (h = 1, 2, \dots, r), \end{matrix} \end{matrix}

(33)

we find the result:

Theorem 6.

The sequence

{g_{k} (A)}_{k \in N}

satisfies the convolution formula:

\begin{matrix} \begin{matrix} g_{k} (A; α_{1}, α_{2}, \dots, α_{r}) = \\ = \sum_{\begin{matrix} k_{1} + k_{2} + \dots + k_{r} = k \\ 0 \leq k_{i} \leq k \end{matrix}} (\binom{k}{k_{1}, k_{2}, \dots, k_{r}}) p_{1, k_{1}} (A, α_{1}) p_{2, k_{2}} (A, α_{2}) \dots p_{r, k_{r}} (A, α_{r}) . \end{matrix} \end{matrix}

(34)

Furthermore, denoting by I the identity matrix, we can proclaim the theorem:

Theorem 7.

The sequence

{g_{k} (A) : = g_{k} (A; a_{1}, a_{2}, \dots, a_{r})}_{k \in N}

satisfies the linear recurrence relation

\begin{matrix} \begin{matrix} g_{k} (A) + a_{1} [A + (k - 1) I] g_{k - 1} (A) + a_{2} (k - 1) [2 A + (k - 2) I] g_{k - 2} (A) + \dots \\ + a_{r} (k - 1) (k - 2) \dots (k - r + 1) [r A + (k - r) I] g_{k - r} (A) = 0 . \end{matrix} \end{matrix}

(35)

8. Conclusions

Starting from the results by Chen Zhuoyu and Qi Lan [9], we have shown convolution formulas and linear recurrence relations satisfied by a generating function containing several parameters. This can be used for number sequences (assuming

x = 1

) or polynomial sequences, depending on several parameters. Illustrative examples are shown both in case of second order or high order recurrence relations.

An extension to the case of matrix polynomials is also included.

Author Contributions

The authors claim to have contributed equally and significantly in this paper. Both authors read and approved the final manuscript.

Funding

This research received no external funding.

Acknowledgments

The authors are grateful to the anonymous referee for his careful reading of the manuscript, which permitted to correct the article.

Conflicts of Interest

The authors declare that they have not received funds from any institution and that they have no conflict of interest.

References

Srivastava, H.M.; Manocha, H.L. A Treatise on Generating Functions; Halsted Press (Ellis Horwood Limited): Chichester, UK; John Wiley and Sons: New York, NY, USA; Chichester, UK; Brisbane, Australia; Toronto, ON, Canada, 1984. [Google Scholar]
Kuş, S.; Tuglu, N.; Kim, T. Bernoulli F-polynomials and Fibo-Bernoulli matrices. Adv. Differ. Equ. 2019, 2019, 145. [Google Scholar] [CrossRef]
Zhang, Y.; Chen, Z. A New Identity Involving the Chebyshev Polynomials. Mathematics 2018, 6, 244. [Google Scholar] [CrossRef]
Ma, Y.; Zhang, W. Some Identities Involving Fibonacci Polynomials and Fibonacci Numbers. Mathematics 2018, 6, 334. [Google Scholar] [CrossRef]
Shen, S.; Chen, L. Some Types of Identities Involving the Legendre Polynomials. Mathematics 2019, 7, 114. [Google Scholar] [CrossRef]
Zhang, W. Some identities involving the Fibonacci numbers and Lucas numbers. Fibonacci Q. 2004, 42, 149–154. [Google Scholar]
Wang, S.Y. Some new identities of Chebyshev polynomials and their applications. Adv. Differ. Equ. 2015, 2015, 335. [Google Scholar]
Kim, T.; Dolgy, D.V.; Kim, D.S.; Seo, J.J. Convolved Fibonacci numbers and their applications. Ars Combin. 2017, 135, 119–131. [Google Scholar]
Chen, Z.; Qi, L. Some convolution formulae related to the second-order linear recurrence sequences. Symmetry 2019, 11, 788. [Google Scholar] [CrossRef]
Sheffer, I.M. Some properties of polynomials sets of zero type. Duke Math. J. 1939, 5, 590–622. [Google Scholar] [CrossRef]
Dattoli, G. Hermite-Bessel and Laguerre-Bessel functions: A by-product of the monomiality principle. In Advanced Special Functions and Applications, Proceedings of the Melfi School on Advanced Topics in Mathematics and Physics, Melfi, Italy, 9–12 May 1999; Cocolicchio, D., Dattoli, G., Srivastava, H.M., Eds.; Aracne Editrice: Rome, Italy, 2000; pp. 147–164. [Google Scholar]
Dattoli, G.; Ricci, P.E.; Srivastava, H.M. (Eds.) Advanced Special Functions and Related Topics in Probability and in Differential Equations. In Proceedings of the Melfi School on Advanced Topics in Mathematics and Physics, Melfi, Italy, 24–29 June 2001. In Appl. Math. Comput. 2003, 141, 1–230. [Google Scholar]
Ben Cheikh, Y. Some results on quasi-monomiality. Appl. Math. Comput. 2003, 141, 63–76. [Google Scholar] [CrossRef]
Boas, R.P.; Buck, R.C. Polynomial Expansions of Analytic Functions; Springer: Berlin/Heidelberg, Germany; Gottingen, Germany; New York, NY, USA, 1958. [Google Scholar]
Sinha, S.K. On a polynomial associated with Gegenbauer polynomial. Proc. Nat. Acad. Sci. India Sect. A 1989, 54, 439–455. [Google Scholar]
Koshy, T. Fibonacci and Lucas Numbers with Applications; Wiley: New York, NY, USA, 2001. [Google Scholar]
Lidl, R. Tschebyscheffpolynome in mehreren variabelen. J. Reine Angew. Math. 1975, 273, 178–198. [Google Scholar]
Ricci, P.E. I polinomi di Tchebycheff in più variabili. Rend. Mat. (Ser. 6) 1978, 11, 295–327. [Google Scholar]
Dunn, K.B.; Lidl, R. Multi-dimensional generalizations of the Chebyshev polynomials. I, II. Proc. Jpn. Acad. 1980, 56, 154–165. [Google Scholar] [CrossRef]
Bruschi, M.; Ricci, P.E. I polinomi di Lucas e di Tchebycheff in più variabili. Rend. Mat. (Ser. 6) 1980, 13, 507–530. [Google Scholar]
Goh, W.; He, M.X.; Ricci, P.E. On the universal zero attractor of the Tribonacci-related polynomials. Calcolo 2009, 46, 95–129. [Google Scholar] [CrossRef]
Srivastava, H.M.; Khan, W.A.; Hiba, H. Some expansions for a class of generalized Humbert Matrix polynomials. Rev. R. Acad. Cienc. Exactas Fís. Nat. Ser. A. Matem. 2019. To appear. [Google Scholar] [CrossRef] [Green Version]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Ricci, P.E.; Natalini, P. General Linear Recurrence Sequences and Their Convolution Formulas. Axioms 2019, 8, 132. https://doi.org/10.3390/axioms8040132

AMA Style

Ricci PE, Natalini P. General Linear Recurrence Sequences and Their Convolution Formulas. Axioms. 2019; 8(4):132. https://doi.org/10.3390/axioms8040132

Chicago/Turabian Style

Ricci, Paolo Emilio, and Pierpaolo Natalini. 2019. "General Linear Recurrence Sequences and Their Convolution Formulas" Axioms 8, no. 4: 132. https://doi.org/10.3390/axioms8040132

APA Style

Ricci, P. E., & Natalini, P. (2019). General Linear Recurrence Sequences and Their Convolution Formulas. Axioms, 8(4), 132. https://doi.org/10.3390/axioms8040132

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

Article Menu

General Linear Recurrence Sequences and Their Convolution Formulas

Abstract

1. Introduction

2. Generating Functions

3. Recurrence Relation

3.1. Properties of the Basic Generating Function

3.1.1. Differential Equation

3.1.2. Differential Identity

3.2. Extension by Convolution

4. The General Case

5. The General Recurrence Relation

Extension to the General Case

6. Illustrative Examples—Second Order Recurrences

Illustrative Examples—Higher Order Recurrences

7. Extension to Matrix Polynomials

8. Conclusions

Author Contributions

Funding

Acknowledgments

Conflicts of Interest

References

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI