Parafunctions of triangular matrices and m-ary partitions of numbers

Using the machinery of paradeterminants and parapermanents developed in [2] we get new relations for some number-theoretical functions natural argument that were studied in [3].

Saved in:
Bibliographic Details
Published in:Algebra and Discrete Mathematics
Date:2016
Main Authors: Stefluk, S., Zatorsky, R.
Format: Article
Language:English
Published: Інститут прикладної математики і механіки НАН України 2016
Online Access:https://nasplib.isofts.kiev.ua/handle/123456789/155197
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Cite this:Parafunctions of triangular matrices and m-ary partitions of numbers / S. Stefluk, R. Zatorsky // Algebra and Discrete Mathematics. — 2016. — Vol. 21, № 1. — С. 144-152. — Бібліогр.: 7 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
_version_ 1860139398187712512
author Stefluk, S.
Zatorsky, R.
author_facet Stefluk, S.
Zatorsky, R.
citation_txt Parafunctions of triangular matrices and m-ary partitions of numbers / S. Stefluk, R. Zatorsky // Algebra and Discrete Mathematics. — 2016. — Vol. 21, № 1. — С. 144-152. — Бібліогр.: 7 назв. — англ.
collection DSpace DC
container_title Algebra and Discrete Mathematics
description Using the machinery of paradeterminants and parapermanents developed in [2] we get new relations for some number-theoretical functions natural argument that were studied in [3].
first_indexed 2025-12-07T17:48:47Z
format Article
fulltext Algebra and Discrete Mathematics RESEARCH ARTICLE Volume 21 (2016). Number 1, pp. 144–152 © Journal “Algebra and Discrete Mathematics” Parafunctions of triangular matrices and m-ary partitions of numbers Svitlana Stefluk and Roman Zatorsky Communicated by R. I. Grigorchuk Abstract. Using the machinavy of paradeterminants and parapermanents developed in [2] we get new relations for some number-theoretical functions natural argument that were studied in [3]. Introduction Partition polynomials together with corresponding linear recurrent equations appear in different areas of mathematics. Therefore, it is im- portant to develop the general theory of partition polynomials, which would unify the results obtained in this area of mathematics. One of these general approaches to studying partition polynomials and its correspond- ing linear recurrent equations is their study with the help of triangular matrices (tables) machinery [1, 2]. The present paper continues the study of properties and interrelations of three number-theoretical functions of a natural argument, which was started in [3]. These functions are the functions bm(n), ξm(n), ([3], p.68- 69.) respectively generalizing the number p(n) of unordered partitions of a positive integer n into positive integer summands and the sum of divisors of a positive integer σ(n), as well as the function dm(n), which for m = 2 equals (−1)t(n), where t(n) is the n-th term of the Prouhet-Thue-Morse 2010 MSC: 12E10. Key words and phrases: partition polynomials, triangular matrices, paradeter- minant, parapermanent, m-ary partition. S. Stefluk, R. Zatorsky 145 sequence [4]. In [3] the authors apply the methods of combinatorial analysis (generatrix method) and linear algebra. As for us, in order to study these functions, we use the general theory of partition polynomials developed with the help of triangular matrix calculus machinery developed by the first autor. At that, we received new relations between these functions and all the proofs are considerably simplified. As the result we get a seweral new relations between functions bm(n), ξm(n) and dm(n) and express them via paradeterminants and parapermanents. 1. Preliminaries This section includes some necessary notions and their properties, which will be needed in the next section. 1.1. Some notions and results concerning triangular matrices In this section we provide basic notions and results about parade- terminants and parapermanents that will be used for the proving of the main results of the paper. Let K be some number field. Definition 1 ([1, 2]). A triangular table A =       a11 a21 a22 ... ... . . . an1 an2 · · · ann       n (1) of numbers from a number field K is called a triangular matrix, an element a11 is an upper element of this triangular matrix, and a number n is its order. Definition 2 ([1, 2]). If A is the triangular matrix (1), then its parade- terminant and parapermanent are the following numbers, respectively: ddet(A) = n ∑ r=1 ∑ p1+...+pr=n (−1)n−r r ∏ s=1 {ap1+...+ps,p1+...+ps−1+1}, (2) pper(A) = n ∑ r=1 ∑ p1+...+pr=n r ∏ s=1 {ap1+...+ps,p1+...+ps−1+1}, 146 Parafunctions and m-ary partitions where the summation is over the set of natural solutions of the equality p1 + . . . + pr = n and bij = {aij} = i ∏ k=j aik, 1 6 j 6 i 6 n. For a parapermanent and paradeterminant of a matrix we will use notations shown in (15) and (16) respectively. Definition 3 ([1,2]). To each element aij of the given triangular matrix (1) we correspond a triangular matrix with this element in the bottom left corner, which we call a corner of the given triangular matrix and denote by Rij(A). It is obvious that the corner Rij(A) is a triangular matrix of order (i − j + 1). The corner Rij(A) includes only those elements ars of the triangular matrix (1), the indices of which satisfy the relations j 6 s 6 r 6 i. Sometimes we extend the range of indeces in (1) from 1,. . . ,n to 0,1,. . . ,n+1 and agree that ddet(R01(A)) = ddet(Rn,n+1(A)) = pper(R01(A)) = pper(Rn,n+1(A)) = 1. (3) When finding values of the paradeterminant and the parapermanent of triangular matrices, it is convenient to use algebraic complements. Definition 4 ([1, 2]). Algebraic complements Dij , Pij to a factorial prod- uct {aij} of a key element aij of the matrix (1) are, respectively, numbers Dij = (−1)i+j · ddet(Rj−1,1) · ddet(Rn,i+1), (4) Pij = pper(Rj−1,1) · pper(Rn,i+1), (5) where Rj−1,1 and Rn,i+1 are corners of the triangular matrix (1). Theorem 1 ([1,2]). If A is the triangular matrix (1), then the parafunc- tions of this matrix can be decomposed by the elements of the last row. With that, the following equalities hold: S. Stefluk, R. Zatorsky 147 ddet(A) = n ∑ s=1 {ans} Dns = n ∑ s=1 (−1)n+s {ans} · ddet(Rs−1,1), (6) pper(A) = n ∑ s=1 {ans} Pns = n ∑ s=1 {ans} · pper(Rs−1,1), (7) where bij = {aij} = i ∏ k=j aik, 1 6 j 6 i 6 n. Theorem 2 (Relation between parapermanent and paradeterminant[1, 2]). If A is the triangular matrix (1), then the following relation holds pper (aij)16j6i6n = ddet ( (−1)δij+1 aij ) 16j6i6n . (8) Corollary 1. For any triangular matrix (bij)16j6i6n, the following equal- ity holds ddet(bij)16j6i6n = pper((−1)δij+1bij)16j6i6n. Theorem 3 ([5]). The following is true ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ a11 a1 0 . . . 0 0 0 a21 a22 a2 . . . 0 0 0 a31 a32 a33 . . . 0 0 0 ... . . . . . . . . . . . . . . . ... an−2,1 an−2,2 an−2,3 . . . an−2,n−2 an−2 0 an−1,1 an−1,2 an−1,3 . . . an−1,n−2 an−1,n−1 an−1 an1 an2 an3 . . . an,n−2 an,n−1 ann ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ ∣ = 〈 a11 a1 a21 a22 a22 a1 a31 a32 a2 a32 a33 a33 ... . . . . . . . . . a1 an−2,1 an−2,2 a2 an−2,2 an−2,3 a3 an−2,3 an−2,4 . . . an−2,n−2 a1 an−1,1 an−1,2 a2 an−1,2 an−1,3 a3 an−1,3 an−1,4 . . . an−2 an−1,n−2 an−1,n−1 an−1,n−1 a1 an1 an2 a2 an2 an3 a3 an3 an4 . . . an−2 an,n−2 an,n−1 an−1 an,n−1 ann ann 〉 (9) 148 Parafunctions and m-ary partitions 1.2. Some data from the general theory of partition polynomials We will need also the following three results. Theorem 4 ([2], Theorem 2.5.3). The following three equalities are equipotent: An = 〈 x(1) x(2) x(1) x(1) ... . . . . . . ·x(n−1) x(n−2) x(n−2) x(n−3) . . . x(1) x(n) x(n−1) x(n−1) x(n−2) . . . x(2) x(1) x(1) 〉 , An = x1An−1 − x2An−2 + . . . + (−1)n−1xnA0, A0 = 1, An = ∑ λ1+2λ2+...+nλn=n (−1)n−k k! λ1! · . . . · λn! xλ1 1 · . . . xλn n , k = λ1 + . . . + λn. Theorem 5 ([7]). Let polynomials yn(x1, x2, . . . , xn), n = 0, 1, . . ., satisfy the recurrence relation yn = x1yn−1 − x2yn−2 + . . . + (−1)n−2xn−1y1 + (−1)n−1anxny0, (10) where y0 = 1. Then the relations yn = ddet       a1x1 a2 x2 x1 x1 ... . . . . . . an xn xn−1 . . . x2 x1 x1       , (11) yn = ∑ λ1+2λ2+...+nλn=n (−1)n−k ( n ∑ i=1 λiai ) (k − 1)! λ1!λ2!·. . .·λn! xλ1 1 xλ2 2 · . . . · xλn n , (12) where k = λ1 + λ2 + . . . + λn, hold. Theorem 6 ([2], Theorem 3.6.1). The following formulae of inversion of partition polynomials written as parafunctions of triangular matrices are valid: S. Stefluk, R. Zatorsky 149 1) bi = 〈 τsr as−r+1 as−r 〉 16r6s6i , (13) ai = 〈 τ−1 s,s−r+1 bs−r+1 bs−r 〉 16r6s6i , i = 1, 2, . . . ; (14) 2) bi = [ τsr as−r+1 as−r ] 16r6s6i , ai = (−1)i−1 〈 τ−1 s,s−r+1 bs−r+1 bs−r 〉 16r6s6i , i = 1, 2, . . . , where ai, bi are arbitrary real variables, τrs are rational numbers. 2. Parafunctions of triangular matrices and m-ary parti- tions of numbers Our first theorem show how functions bm(n), ξm(n), dm(n), studied in [3] can be expressed via paradeterminant and parapermanent. Theorem 7. The following equalities hold: bm(n) =           ξm(1) ξm(2) ξm(1) 1 2ξm(1) ... . . . . . . ξm(n−1) ξm(n−2) ξm(n−2) ξm(n−3) . . . 1 n−1ξm(1) ξm(n) ξm(n−1) ξm(n−1) ξm(n−2) . . . ξm(2) ξm(1) 1 n ξm(1)           , (15) dm(n) = (−1)n 〈 ξm(1) ξm(2) ξm(1) 1 2ξm(1) ... . . . . . . ξm(n−1) ξm(n−2) ξm(n−2) ξm(n−3) . . . 1 n−1ξm(1) ξm(n) ξm(n−1) ξm(n−1) ξm(n−2) . . . ξm(2) ξm(1) 1 n ξm(1) 〉 , (16) ξm(n) = (−1)n−1 〈 bm(1) 2 · bm(2) bm(1) bm(1) ... . . . . . . (n−1)· bm(n−1) bm(n−2) bm(n−2) bm(n−3) . . . bm(1) n· bm(n) bm(n−1) bm(n−1) bm(n−2) . . . bm(2) bm(1) bm(1) 〉 , (17) 150 Parafunctions and m-ary partitions ξm(n) = (−1)n 〈 dm(1) 2 · dm(2) dm(1) dm(1) ... . . . . . . (n−1)· dm(n−1) dm(n−2) dm(n−2) dm(n−3) . . . dm(1) n· dm(n) dm(n−1) dm(n−1) dm(n−2) . . . dm(2) dm(1) dm(1) 〉 , (18) bm(n) = (−1)n 〈 dm(1) dm(2) dm(1) dm(1) ... . . . . . . dm(n−1) dm(n−2) dm(n−2) dm(n−3) . . . dm(1) dm(n) dm(n−1) dm(n−1) dm(n−2) . . . dm(2) dm(1) dm(1) 〉 , (19) dm(n) = (−1)n 〈 bm(1) bm(2) bm(1) bm(1) ... . . . . . . · bm(n−1) bm(n−2) bm(n−2) bm(n−3) . . . bm(1) bm(n) bm(n−1) bm(n−1) bm(n−2) . . . bm(2) bm(1) bm(1) 〉 . (20) Proof. Relations (15), (16) follows from recurrent relations of Theorem 3 (from [3], p. 70). Indeed, each of these equalities is a result of expansion of the paradeterminants on the right side of (15) or (16) by elements of the last raw. Relations (17), (18) can be obtained by inversion of (15), (16) using Theorem 6; (19), (20) follows directly from Theorem 2 in [3], p. 69, and the above Theorem 3 on the relation between paradeterminants and determinants. The following theorem gives recurrent relations between functions bm(n), ξm(n), dm(n). Theorem 8. The following equalities hold: ξm(n) = − ( bm(1)ξm(n − 1) + bm(2)ξm(n − 2) + . . . + bm(n − 1)ξm(1) − nbm(n)ξm(0) ) , (21) ξm(n) = − ( dm(1)ξm(n − 1) + dm(2)ξm(n − 2) + . . . + dm(n − 1)ξm(1) + ndm(n)ξm(0) ) , (22) S. Stefluk, R. Zatorsky 151 bm(n) = − ( dm(1)bm(n − 1) + dm(2)bm(n − 2) + . . . + dm(n − 1)bm(1) + dm(n)bm(0) ) , (23) dm(n) = − ( bm(1)dm(n − 1) + bm(2)dm(n − 2) + . . . + bm(n − 1)dm(1) + bm(n)dm(0) ) , (24) where bm(0) = dm(0) = ξm(0) = 1. Proof. To prove (21) multiply both sides of (17) by (−1)n−1 and expand paradeterminant on the right side of the obtained equality by elements of the last row. As the result, we get (−1)n−1ξm(n) = bm(1)(−1)n−2ξm(n − 1) − bm(2)(−1)n−3ξm(n − 2) + . . . + (−1)n−2bm(n − 1)(−1)0ξm(n − (n − 1)) + (−1)n−1bm(n)(−1)−1ξm(n − n) and hence (21). Similarly, one can prove the relation (22) using (18). Relations (23), (24) can be obtained from (19) and (20) respectively. Let us prove, for example, (23). Multiply both sides of (19) by (−1)n and expand paradeterminant on the right side of obtained equality by elements of the last row. Then (−1)nbm(n) = dm(1)(−1)n−1bm(n − 1) − dm(2)(−1)n−2bm(n − 2) + . . . + (−1)n−2dm(n − 1)(−1)1)bm(n − (n − 1)) + (−1)n−1dm(n)(−1)0bm(n − n), and the required relation follow immediately. In the next theorem, we describe partition polynomials as defined in [6] presenting m-ary numbers bm(n), ξm(n), dm(n). Theorem 9. The following equalities hold: dm(n) = ∑ λ1+2λ2+...+nλn=n (−1)k ξλ1 m (1) · . . . · ξλn m λ1! · . . . · λn!1λ1 · . . . · nλn , (25) ξm(n) = ∑ λ1+2λ2+...+nλn=n (−1)k−1 n(k − 1)! λ1! · . . . · λn! · bλ1 m (1) · . . . · bλn m (n), (26) ξm(n) = ∑ λ1+2λ2+...+nλn=n (−1)k n(k − 1)! λ1! · . . . · λn! · dλ1 m (1) · . . . · dλn m (n), (27) 152 Parafunctions and m-ary partitions bm(n) = ∑ λ1+2λ2+...+nλn=n (−1)k k! λ1! · . . . · λn! · dλ1 m (1) · . . . · dλn m (n), (28) dm(n) = ∑ λ1+2λ2+...+nλn=n (−1)k k! λ1! · . . . · λn! · bλ1 m (1) · . . . · bλn m (n), (29) where k = λ1 + λ2 + . . . + λn. Proof. Partition polynomial corresponding to parapermanent (15), were described by Kachi and Tzermias [3, Theorem 1, p. 68]. Paradeterminant of the same matrix corresponds to the partition polynomial that differs from the previous one only by sign (−1)n−k. Thus (25) holds. The relations for partition polynomials (26), (27) and (28), (29) follow directly from theorems 5 and 4 respectively. References [1] Zatorsky R.A. Theory of paradeterminants and its applications // Algebra and Discrete Mathematics №1, 2007, pp. 109-138. [2] Zatorsky R.A. Calculus of Triangular Matrices and Its Applications. Ivano- Frankivsk, Simyk, 2010, 508 p. (in Ukrainian). [3] Yasuyuki Kachi and Pavlos Tzermias, On the m-ary partition numbers, Algebra and Discrete Mathematics 19, 1 (2015), 67–76. [4] J.-P. Allouche, J. Shallit, The ubiquitous Prouhet-Thue-Morse sequence, In: C. Ding et al. (eds.), Sequences and their Aplications, Proceedings of SETA ’98, Springer- Verlag London, 1999, 1–16. [5] R. A. Zatorsky, Researching of Hessenbergs matrix functions, Carpathian Mathe- matical Publications 3, 1 (2011), 49–55. (in Ukrainian) [6] Riordan J. Combinatorial identities. New York: Wiley, 1968, 256 p. [7] R. Zatorsky, S. Stefluk, On one class of partition polynomials, Algebra and Discrete Mathematics 16, 1 (2013), 127–133. Contact information S. Stefluk, R. Zatorsky Department of Mathematics and Computer Science, Precarpathian Vasyl Stefanyk National University 57 Shevchenka Str. Ivano-Frankivsk, 76025 Ukraine E-Mail(s): ljanys@mail.ru, romazatorsky@gmail.com Web-page(s): www.romaz.pu.if.ua Received by the editors: 24.09.2015 and in final form 19.03.2016.
id nasplib_isofts_kiev_ua-123456789-155197
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
issn 1726-3255
language English
last_indexed 2025-12-07T17:48:47Z
publishDate 2016
publisher Інститут прикладної математики і механіки НАН України
record_format dspace
spelling Stefluk, S.
Zatorsky, R.
2019-06-16T10:47:56Z
2019-06-16T10:47:56Z
2016
Parafunctions of triangular matrices and m-ary partitions of numbers / S. Stefluk, R. Zatorsky // Algebra and Discrete Mathematics. — 2016. — Vol. 21, № 1. — С. 144-152. — Бібліогр.: 7 назв. — англ.
1726-3255
2010 MSC:12E10.
https://nasplib.isofts.kiev.ua/handle/123456789/155197
Using the machinery of paradeterminants and parapermanents developed in [2] we get new relations for some number-theoretical functions natural argument that were studied in [3].
en
Інститут прикладної математики і механіки НАН України
Algebra and Discrete Mathematics
Parafunctions of triangular matrices and m-ary partitions of numbers
Article
published earlier
spellingShingle Parafunctions of triangular matrices and m-ary partitions of numbers
Stefluk, S.
Zatorsky, R.
title Parafunctions of triangular matrices and m-ary partitions of numbers
title_full Parafunctions of triangular matrices and m-ary partitions of numbers
title_fullStr Parafunctions of triangular matrices and m-ary partitions of numbers
title_full_unstemmed Parafunctions of triangular matrices and m-ary partitions of numbers
title_short Parafunctions of triangular matrices and m-ary partitions of numbers
title_sort parafunctions of triangular matrices and m-ary partitions of numbers
url https://nasplib.isofts.kiev.ua/handle/123456789/155197
work_keys_str_mv AT stefluks parafunctionsoftriangularmatricesandmarypartitionsofnumbers
AT zatorskyr parafunctionsoftriangularmatricesandmarypartitionsofnumbers