Next Article in Journal
Optimal Subsidies for Green Products: A Maximal Policy Benefit Perspective
Next Article in Special Issue
A Modified PML Acoustic Wave Equation
Previous Article in Journal
Common Fixed Point Results for Fuzzy Mappings on Complex-Valued Metric Spaces with Homotopy Results
Previous Article in Special Issue
Representation by Chebyshev Polynomials for Sums of Finite Products of Chebyshev Polynomials
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

On the Catalan Numbers and Some of Their Identities

1
School of Mathematics and Statistics, Kashgar University, Xinjiang 844006, China
2
School of Mathematics, Northwest University, Xi’an 710127, Shaanxi, China
*
Author to whom correspondence should be addressed.
Symmetry 2019, 11(1), 62; https://doi.org/10.3390/sym11010062
Submission received: 5 December 2018 / Revised: 31 December 2018 / Accepted: 4 January 2019 / Published: 8 January 2019
(This article belongs to the Special Issue Current Trends in Symmetric Polynomials with their Applications)

Abstract

:
The main purpose of this paper is using the elementary and combinatorial methods to study the properties of the Catalan numbers, and give two new identities for them. In order to do this, we first introduce two new recursive sequences, then with the help of these sequences, we obtained the identities for the convolution involving the Catalan numbers.

1. Introduction

For any non-negative integer n, the famous Catalan numbers C n are defined as C n = 1 n + 1 · 2 n n . For example, the first several Catalan numbers are C 0 = 1 , C 1 = 1 , C 2 = 2 , C 3 = 5 , C 4 = 14 , C 5 = 42 , C 6 = 132 , C 7 = 429 , C 8 = 1430 , ⋯. The Catalan numbers C n satisfy the recursive formula
C n = i = 1 n C i 1 · C i .
The generating function of the Catalan numbers C n is
2 1 + 1 4 x = n = 0 2 n n n + 1 · x n = n = 0 C n · x n .
These numbers occupy a pivotal position in combinatorial mathematics, as many counting problems are closely related to Catalan numbers, and some famous examples can be found in R. P. Stanley [1]. Many papers related to the Catalan numbers and other special sequences can also be found in references [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20], especially the works of T. Kim et al. give a series of new identities for the Catalan numbers, see [9,10,11,12,13,14], these are important results in the related field.
The main purpose of this paper is to consider the calculating problem of the following convolution sums involving the Catalan numbers:
a 1 + a 2 + + a h = n C a 1 · C a 2 · C a 3 C a h ,
where the summation is taken over all h-dimension non-negative integer coordinates ( a 1 , a 2 , , a h ) such that the equation a 1 + a 2 + + a h = n .
About the convolution sums (2), it seems that none had studied it yet, at least we have not seen any related results before. We think this problem is meaningful. The reason is based on the following two aspects: First, it can reveal the profound properties of the Catalan numbers themselves. Second, for the other sequences, such as Fibonacci numbers, Fubini numbers, and Euler numbers, etc. (see [21,22,23]), there are corresponding results, so the Catalan numbers should have a corresponding identity. In this paper, we use the elementary and combinatorial methods to answer this question. That is, we shall prove the following:
Theorem 1.
For any positive integer h, we have the identity
a 1 + a 2 + + a 2 h + 1 = n C a 1 · C a 2 · C a 3 C a 2 h + 1 = 1 ( 2 h ) ! i = 0 h C ( h , i ) j = 0 min ( n , i ) ( n j + h + i ) ! · C n j + h + i ( n j ) ! · i j · ( 4 ) j ,
where C ( h , i ) are defined as C ( 1 , 0 ) = 2 , C ( h , h ) = 1 , C ( h + 1 , h ) = C ( h , h 1 ) ( 8 h + 2 ) · C ( h , h ) , C ( h + 1 , 0 ) = 8 · C ( h , 1 ) 2 · C ( h , 0 ) , and for all integers 1 i h 1 , we have the recursive formula
C ( h + 1 , i ) = C ( h , i 1 ) ( 8 i + 2 ) · C ( h , i ) + ( 4 i + 4 ) ( 4 i + 2 ) · C ( h , i + 1 ) .
Theorem 2.
For any positive integer h and non-negative n, we have
a 1 + a 2 + + a 2 h = n C a 1 · C a 2 · C a 3 C a 2 h = 1 ( 2 h 1 ) ! i = 0 h 1 j = 0 n D ( h , i + 1 ) · i + 1 2 j · ( 4 ) j · ( n j + h + i ) ! · C n j + h + i ( n j ) ! ,
where n + 1 2 i = n + 1 2 · n 1 + 1 2 n i + 1 + 1 2 / i ! , D ( k , i ) are defined as D ( k , 0 ) = 0 , D ( k , k ) = 1 , D ( k + 1 , k ) = D ( k , k 1 ) ( 8 k 2 ) , D ( k + 1 , 1 ) = 24 D ( k , 2 ) 6 D ( k , 1 ) , and for all integers 1 i k 1 ,
D ( k + 1 , i ) = D ( k , i 1 ) ( 8 i 2 ) · D ( k , i ) + 4 i ( 4 i + 2 ) · D ( k , i + 1 ) .
To better illustrate the sequence { C ( k , i ) } and D ( h , i ) , we compute them using mathematical software and list some values in the following Table 1 and Table 2.
Observing these two tables, we can easily find that if 2 k 1 = p is a prime, then for all integers 0 i < k , we have the congruences C k , i 0 mod ( 2 k 1 ) ( 2 k ) and D k , i 0 mod ( 2 k 1 ) ( 2 k 2 ) . So we propose the following two conjectures:
Conjecture 1.
Let p be a prime. Then for any integer 0 i < p + 1 2 , we have the congruence
C p + 1 2 , i 0 mod p ( p + 1 ) .
Conjecture 2.
Let p be a prime. Then for any integer 0 i < p + 1 2 , we have the congruence
D p + 1 2 , i 0 mod p ( p 1 ) .
For some special integers n and h, from Theorem 1 and Theorem 2 we can also deduce several interesting corollaries. In fact if we take n = 0 and h = 1 in the theorems respectively, then we have the following four corollaries:
Corollary 1.
For any positive integer h, we have the identity
i = 0 h C ( h , i ) · ( h + i ) ! · C h + i = ( 2 h ) ! .
Corollary 2.
For any positive integer h, we have the identity
i = 1 h D ( h , i ) · ( h + i 1 ) ! · C h + i 1 = ( 2 h 1 ) ! .
Corollary 3.
For any integer n 0 , we have the identity
a + b + d = n C a · C b · C d = ( n + 1 ) · 1 2 · ( n + 2 ) · C n + 2 ( 2 n + 1 ) · C n + 1 .
Corollary 4.
For any integer n 0 , we have the identity
u + v + w + x + y = n C u · C v · C w · C x · C y = ( n + 1 ) ( n + 2 ) ( 4 n 2 + 8 n + 3 ) 6 · C n + 2 ( n + 3 ) ( n + 2 ) ( n + 1 ) ( 2 n + 3 ) 6 · C n + 3 + ( n + 4 ) ( n + 3 ) ( n + 2 ) ( n + 1 ) 24 · C n + 4 .

2. Several Simple Lemmas

To prove our theorems, we need following four simple lemmas. First we have:
Lemma 1.
Let function f ( x ) = 2 1 + 1 4 x . Then for any positive integer h, we have the identity
( 2 h ) ! · f 2 h + 1 ( x ) = i = 0 h C ( h , i ) · ( 1 4 x ) i · f ( h + i ) ( x ) ,
where f ( i ) ( x ) denotes the i-order derivative of f ( x ) for x, and { C ( h , i ) } are defined as the same as in Theorem 1.
Proof. 
In fact, this identity and its generalization had appeared in D. S. Kim and T. Kim’s important work [9] (see Theorem 3.1), but only in different forms. For the completeness of our results, here we give a different proof by mathematical induction. First from the properties of the derivative we have
f ( x ) = 4 1 + 1 4 x 2 · 1 1 4 x = f 2 ( x ) 1 4 x
or identity
f 2 ( x ) = 1 4 x 1 2 · f ( x ) .
From (3) and note that C ( 1 , 0 ) = 2 and C ( 1 , 1 ) = 1 we have
2 f ( x ) · f ( x ) = 2 1 4 x 1 2 · f ( x ) + 1 4 x 1 2 · f ( x )
and
2 ! f 3 ( x ) = 2 f ( x ) + 1 4 x · f ( x ) = i = 0 1 C ( 1 , i ) · ( 1 4 x ) i · f ( 1 + i ) ( x ) .
That is, Lemma 1 is true for h = 1 .
Assume that Lemma 1 is true for h = k 1 . That is,
( 2 k ) ! · f 2 k + 1 ( x ) = i = 0 k C ( k , i ) · ( 1 4 x ) i · f ( k + i ) ( x ) .
Then from (3), (4), the definition of C ( k , i ) , and the properties of the derivative we can deduce that
( 2 k + 1 ) ! · f 2 k ( x ) · f ( x ) = i = 0 k C ( k , i ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) i = 1 k 4 i · C ( k , i ) · ( 1 4 x ) i 1 · f ( k + i ) ( x )
or
( 2 k + 1 ) ! · f 2 k + 2 ( x ) = i = 0 k C ( k , i ) · ( 1 4 x ) i + 1 2 · f ( k + i + 1 ) ( x ) i = 1 k 4 i · C ( k , i ) · ( 1 4 x ) i 1 2 · f ( k + i ) ( x ) .
Applying (5) and the properties of the derivative we also have
( 2 k + 2 ) ! · f 2 k + 1 ( x ) · f ( x ) = i = 0 k C ( k , i ) · ( 1 4 x ) i + 1 2 · f ( k + i + 2 ) ( x ) i = 0 k 4 i + 2 · C ( k , i ) · ( 1 4 x ) i 1 2 · f ( k + i + 1 ) ( x ) i = 1 k 4 i · C ( k , i ) · ( 1 4 x ) i 1 2 · f ( k + i + 1 ) ( x ) + i = 1 k ( 4 i ) · ( 4 i 2 ) · C ( k , i ) · ( 1 4 x ) i 3 2 · f ( k + i ) ( x )
or note that identity (3) we have
( 2 k + 2 ) ! · f 2 k + 3 ( x ) = i = 0 k C ( k , i ) · ( 1 4 x ) i + 1 · f ( k + i + 2 ) ( x ) i = 0 k 4 i + 2 · C ( k , i ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) i = 1 k 4 i · C ( k , i ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) + i = 1 k ( 4 i ) · ( 4 i 2 ) · C ( k , i ) · ( 1 4 x ) i 1 · f ( k + i ) ( x ) = C ( k , k ) · ( 1 4 x ) k + 1 · f ( 2 k + 2 ) ( x ) + i = 1 k C ( k , i 1 ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) 2 C ( k , 0 ) · f ( k + 1 ) ( x ) i = 1 k 4 i + 2 · C ( k , i ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) i = 1 k 4 i · C ( k , i ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) + 8 · C ( k , 1 ) · f ( k + 1 ) ( x ) + i = 1 k 1 ( 4 i + 4 ) · ( 4 i + 2 ) · C ( k , i + 1 ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) = ( 1 4 x ) k + 1 · f ( 2 k + 2 ) ( x ) + 8 · C ( k , 1 ) 2 · C ( k , 0 ) · f ( k + 1 ) ( x ) + C ( k , k 1 ) ( 8 k + 2 ) · C ( k , k ) · ( 1 4 x ) k · f ( 2 k + 1 ) ( x ) + i = 1 k 1 C ( k , i 1 ) ( 8 i + 2 ) · C ( k , i ) + ( 4 i + 4 ) ( 4 i + 2 ) · C ( k , i + 1 ) × ( 1 4 x ) i · f ( k + i + 1 ) ( x ) = i = 0 k + 1 C ( k + 1 , i ) · ( 1 4 x ) i · f ( k + i + 1 ) ( x ) ,
where we have used the identities C ( k + 1 , k ) = C ( k , k 1 ) ( 8 k + 2 ) · C ( k , k ) , C ( k , k ) = 1 , C ( k + 1 , 0 ) = 8 · C ( k , 1 ) 2 · C ( k , 0 ) and for all integers 1 i k 1 ,
C ( k + 1 , i ) = C ( k , i 1 ) ( 8 i + 2 ) · C ( k , i ) + ( 4 i + 4 ) ( 4 i + 2 ) · C ( k , i + 1 ) .
It is clear that (6) implies Lemma 1 is true for h = k + 1 .
This proves Lemma 1 by mathematical induction. □
Lemma 2.
For any positive integer h, we have the identity
( 2 h 1 ) ! · f 2 h ( x ) = i = 0 h 1 D ( h , i + 1 ) · ( 1 4 x ) i + 1 2 · f ( h + i ) ( x ) ,
where D ( h , i ) are defined as the same as in Theorem 2.
Proof. 
It is clear that using the methods of proving Lemma 1 we can easily deduce Lemma 2. □
Lemma 3.
Let h be any positive integer. Then for any integer k 0 , we have the identity
( 1 4 x ) k · f ( h + k ) ( x ) = n = 0 i = 0 min ( n , k ) C n i + h + k ( n i ) ! k i · ( 4 ) i · x n .
Proof. 
From the binomial theorem we have
( 1 4 x ) k = i = 0 k k i · ( 4 x ) i .
On the other hand, from (1) we also have
f ( h + k ) ( x ) = n = 0 ( n + h + k ) ! · C n + h + k n ! · x n .
Combining (7) and (8) we have
( 1 4 x ) k · f ( h + k ) ( x ) = i = 0 k k i · ( 4 x ) i n = 0 ( n + h + k ) ! · C n + h + k n ! · x n = n = 0 i = 0 k ( n + h + k ) ! · C n + h + k n ! · k i · ( 4 ) i · x n + i = n = 0 i = 0 min ( n , k ) ( n i + h + k ) ! · C n i + h + k ( n i ) ! k i · ( 4 ) i · x n .
This proves Lemma 3. □
Lemma 4.
Let h be any positive integer. Then for any integer k 0 , we have the identity
( 1 4 x ) k + 1 2 · f ( h + k ) ( x ) = n = 0 i = 0 n k + 1 2 i · ( 4 ) i · C n i + h + k ( n i ) ! · x n .
Proof. 
From the power series expansion of the function we know that
( 1 4 x ) k + 1 2 = n = 0 k + 1 2 n · ( 4 ) n · x n .
Applying (8) and (9) we have
( 1 4 x ) k + 1 2 · f ( h + k ) ( x ) = n = 0 k + 1 2 n · ( 4 ) n · x n n = 0 ( n + h + k ) ! · C n + h + k n ! · x n = n = 0 i = 0 n k + 1 2 i · ( 4 ) i · ( n i + h + k ) ! · C n i + h + k ( n i ) ! · x n .
This proves Lemma 4. □

3. Proofs of the Theorems

In this section, we shall complete the proofs of our theorems. First we prove Theorem 1. From (1) and the multiplicative properties of the power series we have
( 2 h ) ! · f 2 h + 1 ( x ) = ( 2 h ) ! n = 0 a 1 + a 2 + + a 2 h + 1 = n C a 1 · C a 2 C a 2 h + 1 · x n .
On the other hand, from Lemma 1 and Lemma 3 we also have
( 2 h ) ! · f 2 h + 1 ( x ) = i = 0 h C ( h , i ) · ( 1 4 x ) i · f ( h + i ) ( x ) = n = 0 i = 0 h C ( h , i ) j = 0 min ( n , i ) ( n j + h + i ) ! C n j + h + i ( n j ) ! i j ( 4 ) j x n .
Combining (10) and (11) we may immediately deduce the identity
a 1 + a 2 + + a 2 h + 1 = n C a 1 · C a 2 · C a 3 C a 2 h + 1 = 1 ( 2 h ) ! i = 0 h C ( h , i ) j = 0 min ( n , i ) ( n j + h + i ) ! · C n j + h + i ( n j ) ! · i j · ( 4 ) j .
This proves Theorem 1.
Now we prove Theorem 2. For any positive integer h, from (1) we have
f 2 h ( x ) = n = 0 a 1 + a 2 + + a 2 h = n C a 1 · C a 2 C a 2 h · x n .
On the other hand, from Lemma 2 and Lemma 4 we also have
( 2 h 1 ) ! · f 2 h ( x ) = i = 0 h 1 D ( h , i + 1 ) · ( 1 4 x ) i + 1 2 · f ( h + i ) ( x ) = i = 0 h 1 D ( h , i + 1 ) n = 0 j = 0 n i + 1 2 j ( 4 ) j ( n j + h + i ) ! · C n j + h + i ( n j ) ! x n = n = 0 i = 0 h 1 j = 0 n D ( h , i + 1 ) i + 1 2 j ( 4 ) j ( n j + h + i ) ! C n j + h + i ( n j ) ! x n .
From (12), (13), and Lemma 2 we may immediately deduce the identity
a 1 + a 2 + + a 2 h = n C a 1 · C a 2 · C a 3 C a 2 h = 1 ( 2 h 1 ) ! i = 0 h 1 j = 0 n D ( h , i + 1 ) · i + 1 2 j · ( 4 ) j · ( n j + h + i ) ! · C n j + h + i ( n j ) ! .
This completes the proof of Theorem 2.

4. Conclusions

The main results of this paper are Theorem 1 and Theorem 2. They gave two special expressions for convolution (2). In addition, Corollary 1 gives a close relationship between C ( h , i ) and C h + i . Corollary 2 gives a close relationship between D ( h , i ) and D h + i 1 . Corollary 3 and Corollary 4 give two exact representations for the special cases of Theorem 1 with h = 1 and h = 2 .
About the new sequences C ( h , i ) and D ( h , i ) , we proposed two interesting conjectures related to congruence mod p , where p is an odd prime. We believe that these conjectures are correct, but at the moment we cannot prove them. We also believe that these two conjectures will certainly attract the interest of many readers, thus further promoting the study of the properties of C ( h , i ) and C h + i .

Author Contributions

All authors have equally contributed to this work. All authors read and approved the final manuscript.

Funding

This work is supported by the National Natural Science Foundation (N. S. F.) (11771351) and (11826205) of China.

Acknowledgments

The authors would like to thank the Editor and referees for their very helpful and detailed comments, which have significantly improved the presentation of this paper.

Conflicts of Interest

The authors declare that there are no conflict of interest regarding the publication of this paper.

References

  1. Stanley, R.P. Enumerative Combinatorics (Vol. 2); Cambridge Studieds in Advanced Mathematics; Cambridge University Press: Cambridge, UK, 1997; p. 49. [Google Scholar]
  2. Chu, W.C. Further identities on Catalan numbers. Discret. Math. 2018, 341, 3159–3164. [Google Scholar] [CrossRef]
  3. Joseph, A.; Lamprou, P. A new interpretation of the Catalan numbers arising in the theory of crystals. J. Algebra 2018, 504, 85–128. [Google Scholar] [CrossRef]
  4. Allen, E.; Gheorghiciuc, I. A weighted interpretation for the super Catalan numbers. J. Integer Seq. 2014, 9, 17. [Google Scholar]
  5. Tauraso, R. qq-Analogs of some congruences involving Catalan numbers. Adv. Appl. Math. 2012, 48, 603–614. [Google Scholar] [CrossRef]
  6. Liu, J.C. Congruences on sums of super Catalan numbers. Results Math. 2018, 73, 73–140. [Google Scholar] [CrossRef]
  7. Qi, F.; Shi, X.T.; Liu, F.F. An integral representation, complete monotonicity, and inequalities of the Catalan numbers. Filomat 2018, 32, 575–587. [Google Scholar] [CrossRef]
  8. Qi, F.; Shi, X.T.; Mahmoud, M. The Catalan numbers: A generalization, an exponential representation, and some properties. J. Comput. Anal. Appl. 2017, 23, 937–944. [Google Scholar]
  9. Kim, D.S.; Kim, T. A new approach to Catalan numbers using differential equations. Russ. J. Math. Phys. 2017, 24, 465–475. [Google Scholar] [CrossRef] [Green Version]
  10. Kim, T.; Kim, D.S. Some identities of Catalan-Daehee polynomials arising from umbral calculus. Appl. Comput. Math. 2017, 16, 177–189. [Google Scholar]
  11. Kim, T.; Kim, D.S. Differential equations associated with Catalan-Daehee numbers and their applications. Revista de la Real Academia de Ciencias Exactas, Físicas y Naturales. Serie A. Matemáticas 2017, 111, 1071–1081. [Google Scholar] [CrossRef]
  12. Kim, D.S.; Kim, T. Triple symmetric identities for w-Catalan polynomials. J. Korean Math. Soc. 2017, 54, 1243–1264. [Google Scholar]
  13. Kim, T.; Kwon, H.-I. Revisit symmetric identities for the λ-Catalan polynomials under the symmetry group of degree n. Proc. Jangjeon Math. Soc. 2016, 19, 711–716. [Google Scholar]
  14. Kim, T. A note on Catalan numbers associated with p-adic integral on Zp. Proc. Jangjeon Math. Soc. 2016, 19, 493–501. [Google Scholar]
  15. Basic, B. On quotients of values of Euler’s function on the Catalan numbers. J. Number Theory 2016, 169, 160–173. [Google Scholar] [CrossRef]
  16. Aker, K.; Gursoy, A.E. A new combinatorial identity for Catalan numbers. Ars Comb. 2017, 135, 391–398. [Google Scholar]
  17. Qi, F.; Guo, B.N. Integral representations of the Catalan numbers and their applications. Mathematics 2017, 5, 40. [Google Scholar] [CrossRef]
  18. Hein, N.; Huang, J. Modular Catalan numbers. Eur. J. Combin. 2017, 61, 197–218. [Google Scholar] [CrossRef] [Green Version]
  19. Dilworth, S.J.; Mane, S.R. Applications of Fuss-Catalan numbers to success runs of Bernoulli trials. J. Probab. Stat. 2016, 2016, 2071582. [Google Scholar] [CrossRef]
  20. Dolgy, D.V.; Jang, G.-W.; Kim, D.S.; Kim, T. Explicit expressions for Catalan-Daehee numbers. Proc. Jangjeon Math. Soc. 2017, 20, 1–9. [Google Scholar]
  21. Zhang, Y.X.; Chen, Z.Y. A new identity involving the Chebyshev polynomials. Mathematics 2018, 6, 244. [Google Scholar] [CrossRef]
  22. Zhao, J.H.; Chen, Z.Y. Some symmetric identities involving Fubini polynomials and Euler numbers. Symmetry 2018, 10, 359. [Google Scholar]
  23. Zhang, W.P. Some identities involving the Fibonacci numbers and Lucas numbers. Fibonacci Q. 2004, 42, 149–154. [Google Scholar]
Table 1. Values of C ( k , i ) .
Table 1. Values of C ( k , i ) .
C ( k , i ) i = 0 i = 1 i = 2 i = 3 i = 4 i = 5 i = 6
k = 1 −21
k = 2 12−121
k = 3 −120180−301
k = 4 1680−3360840−561
k = 5 −30,24075,600−25,2002520−901
k = 6 665,280−1,995,840831,600−110,8805940−1321
Table 2. Values of D ( k , i ) .
Table 2. Values of D ( k , i ) .
D ( k , i ) i = 0 i = 1 i = 2 i = 3 i = 4 i = 5 i = 6
k = 1 01
k = 2 0−61
k = 3 060−201
k = 4 0−840420−421
k = 5 015,120−10,0801512−721
k = 6 0−332,640277,200−55,4403960−1101

Share and Cite

MDPI and ACS Style

Zhang, W.; Chen, L. On the Catalan Numbers and Some of Their Identities. Symmetry 2019, 11, 62. https://doi.org/10.3390/sym11010062

AMA Style

Zhang W, Chen L. On the Catalan Numbers and Some of Their Identities. Symmetry. 2019; 11(1):62. https://doi.org/10.3390/sym11010062

Chicago/Turabian Style

Zhang, Wenpeng, and Li Chen. 2019. "On the Catalan Numbers and Some of Their Identities" Symmetry 11, no. 1: 62. https://doi.org/10.3390/sym11010062

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

Article Metrics

Back to TopTop