Journal for Algebra and Number Theory Academia Volume 4, Issue 3, September 2014, Pages 77-87 © 2014 Mili Publications
SOME INTEGRAL REPRESENTATIONS AND PROPERTIES OF LAH NUMBERS BAI-NI GUO and FENG QI School of Mathematics and Informatics Henan Polytechnic University Jiaozuo City, Henan Province, 454010, China E-mail:
[email protected] [email protected] College of Mathematics Inner Mongolia University for Nationalities Tongliao City, Inner Mongolia Autonomous Region 028043, China and Department of Mathematics College of Science Tianjin Polytechnic University Tianjin City, 300387, China E-mail:
[email protected] [email protected] [email protected] Abstract In the paper, the authors find some integral representations and discover some properties of Lah numbers.
1. Introduction In combinatorics, Lah numbers, discovered by Ivo Lah in 1955 and usually denoted by Ln, k , count the number of ways a set of n elements can 2010 Mathematics Subject Classification: 05A10, 05A19, 05A20, 11B34, 11B37, 11B65, 11B75, 11B83, 11R33, 11Y35, 11Y55, 26A48, 26A51, 33B10, 33C15, 33C20. Keywords: integral representation; convexity; total sum, Lah number; modified Bessel function of the first kind; exponential function; absolutely convex function; absolutely convex sequence; generating function. Received August 17, 2014; Revised September 11, 2014 Accepted September 12, 2014 Academic Editor: Serkan Araci
78
BAI-NI GUO and FENG QI
be partitioned into k nonempty linearly ordered queues and have an explicit formula
n 1 n! L n, k . k 1 k!
(1.1)
Lah numbers Ln, k may also be interpreted as coefficients expressing rising factorials x n in terms of falling factorials x n , where
x x 1 x 2 x n 1, x n
n 1, n 0
(1.2)
x x 1 x 2 x n 1, 1
n 1, n 0.
(1.3)
1
and
x
n
Lah numbers Ln, k may be generated by
1 x k k! 1 x
n
x L n, k . n! n 0
(1.4)
For more information on Lah numbers Ln, k , please refer to [3, p. 156]. In the theory of special functions, it is well known that the modified Bessel function of the first kind I z may be defined [1, p. 375, 9.6.10] by
I z
k 0
2k 1 z k! k 1 2
(1.5)
for and z , where represents the classical Euler gamma function which may be defined [1, p. 255] by
z
0 t
z 1 t
e dt
(1.6)
for Rz 0. In this paper, we will find some integral representations and properties of Lah numbers Ln, k .
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
SOME INTEGRAL REPRESENTATIONS AND PROPERTIES OF … 79 2. Integral Representations of Lah Numbers We first establish integral representations of Lah numbers Ln, k , in which the exponential function e 1 x and the modified Bessel function of the first kind I1 are involved. Theorem 2.1. For 1 m n and x 0, we have n
x
e L n, k x k x n 0 k 1
I1 2 t t n 1 2e t x dt
(2.1)
and
L n, m
1 dm lim I1 2 t t n 1 2 m! x 0 0 dx m
e x t x dt. n x
(2.2)
Proof. In [22, Theorem 1.2], among other things, it was obtained that the function
H k z e1 z
k
1 1 m! z m m0
(2.3)
for k 0 and z 0 has the integral representation
H k z
1 F 1; k 1, k 2; t t k e z t dt k!k 1! 0 1 2
(2.4)
for Rz 0, where p Fq a1 , , a p ; b1 , , bq ; x stands for the generalized hyper-geometric series which may be defined by p Fq a1 ,
, a p ; b1 , , bq ; x
n 0
a1 n a p n x n b1 n bq n n!
(2.5)
for complex numbers ai and bi 0, 1, 2, and for positive integers
p, q . See also [18, Section 1.2] and [19, Lemma 2.1]. When k 0, the integral representation (2.4) becomes e1 z 1
0
I1 2 t zt e dt t
(2.6)
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
80
BAI-NI GUO and FENG QI
for Rz 0. By the way, the integral representation (2.6) has been applied in [5]. Hence, for n and x 0, we have
e1 x n 1n
0
I1 2 t t n 1 2e x t dt.
(2.7)
In [3, Chapter III, Exercise 7, p. 158], [4], and [26, Theorem 2.2], the following explicit formula for computing the n-th derivative of the exponential function e 1 x was inductively obtained:
e 1 x n 1n e 1
x
n
1k L n, k n1 k . x k 1
(2.8)
By the way, the formula (2.8) have been applied in [12, 13, 14, 16, 18, 19, 22]. Combining (2.7) and (2.8) and rearranging yield
e1 x
n
k 1
L n, k
n
k 1
1 x nk
L n, k
1 x
k
0
0
I1 2 t t n 1 2e xt dt,
I1 2 t t n 1 2 x n e xt 1 x dt,
which may be rewritten as (2.1). Differentiating 1 m n times on both sides of (2.1) results in n
L n, k
k m
k! x k m k m!
dm I1 2 t t n 1 2 0 dx m
e x t x dt. n x
Letting x 0 in the above equation leads to (2.2). The proof of Theorem 2.1 is complete.
3. Convexity of Total Sum of Lah Numbers With the help of the integral representation (2.1), we are now in a position to prove the convexity of the total sum of Lah numbers. Theorem 3.1. For n , the total sum of Lah numbers
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
SOME INTEGRAL REPRESENTATIONS AND PROPERTIES OF … 81 n
n
L n, k k 1
(3.1)
is a convex sequence. Proof. Letting x 1 in (2.1) gives
n
0
I1 2 t t n 1 2e 1 t dt.
It is clear that the function t x for t 0 satisfies result, when t 0, the function t
x
d kt x dx
k
t x ln t k . As a
is convex with respect to x, that is,
tx t y t x y 2 , t 0, x , y . 2 Taking in the above inequality x n 1 and y n 1 for n yields
t n 1 t n 1 t n , t 0, n . 2 Consequently, we have 1 n 1 2 1 t I 2 t t e dt I1 2 t t n 21 2e 1 t dt 1 2 0 0
0
I1 2 t t n 11 2e 1 t dt,
that is, n n 2 n 1 , n . 2
Hence, the sequence n is convex. The proof of Theorem 3.1 is complete. 4. Two Recoveries of the Formula (1.1) Finally, as by-product, two recoveries of the formula (1.1) for Lah numbers Ln, k may be carried out as follows.
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
82
BAI-NI GUO and FENG QI
4.1. First Recovery. The generating function (1.4) may be rewritten as k 1k 1 x k! 1 x
n
1n L n, k x . n! nk
(4.1)
The equation (4.1) may be reformulated as
1k 1
1
k! 1 x k
nk
1n L n, k x
nk
n!
n
1n k L n k, k x . n k ! n 0
Because
1
1 x k
1 t k 1 e 1 x t dt, k 1! 0
(4.2)
we have 1 1 t k 1e 1 x t dt k! k 1! 0
1n L n k, k
n 0
xn . n k !
Differentiating m times with respect to x on both sides of the above equation gives 1 t m k 1e 1 x t dt k! k 1! 0
1m 1
1n L n k, k
n m
n! x n m . n m! n k !
Taking x 0 in the above equation yields 1 m! t m k 1e t dt 1m L m k, k m k ! k! k 1! 0
1m 1
which may be rearranged as
L m k, k
m k ! 1 m!
1 t m k 1e t dt k! k 1! 0
m k ! 1 m k 1! m k ! m k 1 . m! k! k 1! k! k 1
The formula (1.1) is thus recovered. 4.2. Second recovery. This recovery, which is simpler than the first one above, is supplied by the anonymous referee. Newton’s binomial series for negative exponents gives us Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
SOME INTEGRAL REPRESENTATIONS AND PROPERTIES OF … 83
1 k
1 x
n 0
k n 1 n x . k 1
Multiplying both sides by x k , followed by a variable shift on the sum, yields
xk
1 x k
k n 1 n k n 1 n n 1 x n . x x n! k 1 k 1 k 1 n! n 0 nk nk
Dividing both sides by k! gives
1 x k k! 1 x
n 1 n! x n , k 1 k! n! nk
which means the formula (1.1). 5. Remarks Remark 5.1. An infinitely differentiable function f on an interval I is called absolutely convex on I if f 2k x 0 on I. See either [11, p. 375, Definition 3], or [17, p. 2731, Definition 4.5], or [23, p. 617, Definition 3], or [24, p. 3356, Definition 3]. A sequence n 0 is said to be absolutely convex if its elements are non-negative and its successive differences satisfy
2kn 0
(5.1)
for n, k 0, where
kn
k
k
1m n k m . m m0
(5.2)
Motivated by Theorem 3.1 and suggested by some results on interpolation by completely monotonic functions in, for example, [9] and [25, pp. 163-166], we conjecture that the total sum n of Lah numbers Ln, k is an absolutely convex sequence. Remark 5.2. In the early morning of 30 December 2013, the authors searched out the paper [4] in which the formula (2.8) was also found Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
84
BAI-NI GUO and FENG QI
independently by five approaches. The motivation of the paper [4] is different from the ones of [26]. The motivations of the formula (2.8) in [26] essentially originated from the articles [6, 8, 10]. For more information, please refer to the expository and survey articles [15, 20, 21]. Remark 5.3. In an-email bearing the date March 7, 2014, Dr. Miloud Mihoubi in Algeria recommended the paper [2] to the authors. In [2, Lemma], it was stated that if
Pm, k x
m
Lk m, n x n , n 1
(5.3)
then the m roots of Pm, k x are real, distinct, and non-positive for all m , where the associated Lah numbers Lk m, n for k 0 may be defined by
Lk m, n
m! n!
n
n
1n r r r 1
m rk 1 m
(5.4)
and Lk m, n 0 for n m. Since L1 m, n Lm, n, see [2, p. 158, Equation (4)], when k 1, the polynomial Pm, k x becomes
Pm, 1 x
m
L m, n x n . n 1
(5.5)
For n , the integer polynomial
n x
n
L n 1, k 1x k k 0
(5.6)
of degree n satisfies
n x
Pn 1, 1 x x
n
k 0
L n 1, k 1 x k
n 1
L n 1, k x k 1 0. k 1
(5.7)
This shows a connection between n x and Pm, k k . Remark 5.4. In [13], an explicit formula for Bell numbers Bn in terms of Stirling numbers of the second kind S n, k together with Lah numbers
Ln, k respectively was established as follows: For n , Bell numbers Bn
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
SOME INTEGRAL REPRESENTATIONS AND PROPERTIES OF … 85 may be expressed as n
Bn
k 1
k
Lk, S n, k. 1
1n k
(5.8)
This equation may be rewritten as n
1n k ak S n, k Bn , k 1
(5.9)
where a k is sequence A000262 in the Online Encyclopedia of Integer Sequences. Such a sequence a k has a nice combinatorial interpretation: it counts “the sets of lists, or the number of partitions of 1, 2, , k into any number of lists, where a list means an ordered subset.” This reveals the combinatorial interpretation of the total sum k of Lah numbers Lk, . Another combinatorial interpretation of the sum k of Lah numbers
Ln, k , pointed out by the referee, states that the total sum k is clearly the number of ways to partition n elements into any number of nonempty lists. Remark 5.5. This paper is a corrected and modified version of the preprint [7]. Acknowledgments The authors appreciate anonymous referees for their careful corrections to and valuable comments on the original version of this paper. References [1]
M. Abramowitz and I. A. Stegun (Eds), Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, National Bureau of Standards, Applied Mathematics Series 55, 10th printing, Dover Publications, New York and Washington, 1972.
[2]
J. C. Ahuja and E. A. Enneking, Concavity property and a recurrence relation for associated Lah numbers, Fibonacci Quart. 17(2) (1979), 158-161.
[3]
L. Comtet, Advanced Combinatorics: The Art of Finite and Infinite Expansions, Revised and Enlarged Edition, D. Reidel Publishing Co., Dordrecht and Boston, 1974.
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
86 [4]
BAI-NI GUO and FENG QI S. Daboul, J. Mangaldan, M. Z. Spivey and P. J. Taylor, The Lah numbers and the n-th derivative of e1
x
, Math. Mag. 86(1) (2013), 39-47.
Available online at http://dx.doi.org/10.4169/math.mag.86.1.039. [5]
B.-N. Guo and F. Qi, An explicit formula for Bell numbers in terms of Stirling numbers and hypergeometric functions, Glob. J. Math. Anal. 2(4) (2014), 243-248. Available online at http://dx.doi.org/10.14419/gjma.v2i4.3310.
[6]
B.-N. Guo and F. Qi, Refinements of lower bounds for polygamma functions, Proc. Amer. Math. Soc. 141(3) (2013), 1007-1015. Available online at http://dx.doi.org/10.1090/S0002-9939-2012-11387-5.
[7]
B.-N. Guo and F. Qi, Some integral representations and properties of Lah numbers, available online at http://arxiv.org/abs/1402.2367.
[8]
B.-N. Guo and F. Qi, Some properties of the psi and polygamma functions, Hacet. J. Math. Stat. 39(2) (2010), 219-231.
[9]
S. Guo, Some properties of completely monotonic sequences and related interpolation, Appl. Math. Comput. 219(10) (2013), 4958-4962. Available online at http://dx.doi.org/10.1016/j.amc.2012.11.073.
[10]
W.-H. Li, F. Qi and B.-N. Guo, On proofs for monotonicity of a function involving the psi and exponential functions, Analysis (Munich) 33(1) (2013), 45-50. Available online at http://dx.doi.org/10.1524/anly.2013.1175.
[11]
D. S. Mitrinović, J. E. Pecarić and A. M. Fink, Classical and New Inequalities in Analysis, Kluwer Academic Publishers, 1993.
[12]
F. Qi, A diagonal recurrence formula for Stirling numbers of the first kind, available online at http://arxiv.org/abs/1310.5920.
[13]
F. Qi, An explicit formula for computing Bell numbers in terms of Lah and Stirling numbers, available online at http://arxiv.org/abs/1401.1625.
[14]
F.
Qi,
Six
proofs
for
an
identity
of
Lah
numbers,
available
online
at
http://arxiv.org/abs/1402.2035. [15]
F. Qi, Bounds for the ratio of two gamma functions, J. Inequal. Appl. 2010 (2010), Article ID 493058, 84 pages; Available online at http://dx.doi.org/10.1155/2010/493058.
[16]
F. Qi, Explicit formulas for computing Bernoulli numbers of the second kind and Stirling numbers of the first kind, Filomat 28(2) (2014), 319-327. Available online at http://dx.doi.org/10.2298/FIL14023190.
[17]
F. Qi, Generalized weighted mean values with two parameters, R. Soc. Lond. Proc. Ser. A Math. Phys. Eng. Sci. 454(1978) (1998), 2723-2732; Available online at http://dx.doi.org/10.1098/rspa.1998.0277.
[18]
F. Qi, Properties of modified Bessel functions and completely monotonic degrees of differences between exponential and trigamma functions, Math. Inequal. Appl. 18 (2015), in press; Available online at http://arxiv.org/abs/1302.6731.
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014
SOME INTEGRAL REPRESENTATIONS AND PROPERTIES OF … 87 [19]
F. Qi and C. Berg, Complete monotonicity of a difference between the exponential and trigamma functions and properties related to a modified Bessel function, Mediterr. J. Math. 10(4) (2013), 1685-1696. Available online at http://dx.doi.org/10.1007/s00009-013-0272-2.
[20]
F. Qi and Q.-M. Luo, Bounds for the ratio of two gamma functions: from Wendel’s asymptotic relation to Elezović-Giordano-Pecarić’s theorem, J. Inequal. Appl. 2013 (2013), 542, 20 pages; Available online at http://dx.doi.org/10.1186/1029-242X-2013-542.
[21]
F. Qi and Q.-M. Luo, Bounds for the ratio of two gamma functions – From Wendel’s and related inequalities to logarithmically completely monotonic functions, Banach J. Math. Anal. 6(2) (2012), 132-158.
[22]
F. Qi and S.-H. Wang, Complete monotonicity, completely monotonic degree, integral representations, and an inequality related to the exponential, trigamma, and modified Bessel functions, Glob. J. Math. Anal. 2(3) (2014), 91-97. Available online at http://dx.doi.org/10.14419/gjma.v2i3.2919.
[23]
F. Qi and S.-L. Xu, Refinements and extensions of an inequality, II, J. Math. Anal. Appl. 211(2) (1997), 616-620. Available online at http://dx.doi.org/10.1006/jmaa.1997.5318.
[24]
F. Qi and S.-L. Xu, The function bx a x x : inequalities and properties, Proc. Amer. Math. Soc. 126(11) (1998), 3355-3359. Available online at http://dx.doi.org/10.1090/S0002-9939-98-04442-6.
[25]
D. V. Widder, The Laplace Transform, Princeton University Press, Princeton, 1946.
[26]
X.-J. Zhang, F. Qi and W.-H. Li, Properties of three functions relating to the exponential function and the existence of partitions of unity, Int. J. Open Probl. Comput. Sci. Math. 5(3) (2012), 122-127.
Journal for Algebra and Number Theory Academia, Volume 4, Issue 3, September 2014