Abstract
Based on the Padé approximation method, we determine the coefficients \(a_{j}\) and \(b_{j}\) (\(1\leq j\leq k\)) such that
where \(k\geq1\) is any given integer. Based on the obtained result, we establish new bounds for the gamma function.
Similar content being viewed by others
1 Introduction
Stirling’s formula
has many applications in statistical physics, probability theory and number theory. Actually, it was first discovered in 1733 by the French mathematician Abraham de Moivre (1667-1754) in the form
when he was studying the Gaussian distribution and the central limit theorem. Afterwards, the Scottish mathematician James Stirling (1692-1770) found the missing constant \(\sqrt{2\pi}\) when he was trying to give the normal approximation of the binomial distribution.
Stirling’s series for the gamma function is given (see [1], p.257, Eq. (6.1.40)) by
as \(x\to\infty\), where \(B_{n}\) \((n \in\mathbb{N}_{0}:=\mathbb{N} \cup\{0\} )\) are the Bernoulli numbers defined by the following generating function:
The following asymptotic formula is due to Laplace:
as \(x\to\infty\) (see [1], p.257, Eq. (6.1.37)).
The expression (1.4) is sometimes incorrectly called Stirling’s series (see [2], pp.2-3). Stirling’s formula is in fact the first approximation to the asymptotic formula (1.4). Stirling’s formula has attracted much interest of many mathematicians and has motivated a large number of research papers concerning various generalizations and improvements (see [3–54] and the references cited therein). It is interesting to note that the aforementioned mathematicians represent many nationalities. So the topic is of interest for mathematicians from diverse cultural background.
Using the Maple software, we find, as \(x\to\infty\),
and
Based on the Padé approximation method, in this paper we develop the approximation formulas (1.5) and (1.6) to produce a general result. More precisely, we determine the coefficients \(a_{j}\) and \(b_{j}\) (\(1\leq j\leq k\)) such that
where \(k\geq1\) is any given integer. Based on the obtained result, we establish new bounds for the gamma function.
The numerical values given in this paper have been calculated via the computer program MAPLE 13.
2 Lemmas
The following lemmas are required in our present investigation.
Lemma 2.1
[9]
Let r be a given nonzero real number. The gamma function has the following asymptotic formula:
with the coefficients \(b_{j}=b_{j}(r) \) (\(j=1,2,\ldots\)) given by
where \(B_{n}\) \((n \in\mathbb{N}_{0}:=\mathbb{N} \cup\{0\} )\) are the Bernoulli numbers defined in (1.3), summed over all nonnegative integers \(k_{j}\) satisfying the equation \(k_{1}+2k_{2}+\cdots+jk_{j}=j\).
Laplace formula (1.4) can be rewritten as
with the coefficients \(c_{j}\) given by
Lemma 2.2
[55], Theorem 8
Let \(n \geq0\) be an integer. The functions
and
are completely monotonic on \((0, \infty)\). Here \(B_{n}\) \((n \in\mathbb{N}_{0}:=\mathbb{N}\cup\{0\})\) are the Bernoulli numbers.
Remark 2.1
Lemma 2.2 can be stated as follows: for every \(m\in\mathbb{N}_{0}\), the function
is completely monotonic on \((0, \infty)\).
In 2006, Koumandos [56] presented a simpler proof of complete monotonicity of the functions \(R_{m}(x)\). In 2009, Koumandos and Pedersen [57], Theorem 2.1, strengthened this result.
From \(F_{n}'(x)<0\) and \(G_{n}'(x)<0\) for \(x>0\), we obtain
where \(\psi(x)=\Gamma'(x)/\Gamma(x)\) is the psi (or digamma) function. Noting that
holds, we obtain from (2.5) that for \(x>0\),
3 Approximations to the gamma function
For our later use, we introduce Padé approximant (see [58–61]). Let f be a formal power series
The Padé approximation of order \((p, q)\) of the function f is the rational function, denoted by
where \(p\geq0\) and \(q\geq1\) are two given integers, the coefficients \(a_{j}\) and \(b_{j}\) are given by (see [58–60])
and the following holds:
Thus, the first \(p + q + 1\) coefficients of the series expansion of \([p/q]_{f}\) are identical to those of f. Moreover, we have (see [61])
with \(f_{n}(x) = c_{0}+ c_{1}x+ \cdots+ c_{n}x^{n}\), the nth partial sum of the series f in (3.1) (\(f_{n}\) is identically zero for \(n < 0\)).
Let
It follows from (2.3) that, as \(x\to\infty\),
with the coefficients \(c_{j}\) given by (2.4). In what follows, the function f is given in (3.6).
Based on the Padé approximation method, we now give a derivation of formula (1.5). To this end, we consider
Noting that
holds, we have, by (3.3),
that is,
We thus obtain that
and we have, by (3.4),
We now give a derivation of formula (1.6). To this end, we consider
Noting that (3.8) holds, we have, by (3.3),
that is,
We thus obtain that
and we have, by (3.4),
From the Padé approximation method and the expansion (3.7), we now present a general result given by Theorem 3.1.
Theorem 3.1
The Padé approximation of order \((p, q)\) of the Laplace asymptotic formula of the function \(f(x)=\frac{\Gamma(x+1)}{\sqrt{2\pi x}(x/e)^{x}}\) (at the point \(x=\infty\)) is the following rational function:
where \(p\geq1\) and \(q\geq1\) are any given integers, the coefficients \(a_{j}\) and \(b_{j}\) are given by
and \(c_{j}\) is given in (2.4), and the following holds:
Moreover, we have
with \(f_{n}(x)=\sum_{j=0}^{n}\frac{c_{j}}{x^{j}}\), the nth partial sum of the asymptotic series (3.7).
Remark 3.1
Using (3.14), we can also derive (3.9) and (3.10). Indeed, we have
and
Setting \((p, q)=(k, k)\) in (3.13), we obtain the following corollary.
Corollary 3.1
As \(x\to\infty\),
where \(k\geq1\) is any given integer, the coefficients \(a_{j}\) and \(b_{j}\) (\(1\leq j\leq k\)) are given by
and \(c_{j}\) is given in (2.4).
Setting \(k=3\) and \(k=4\) in (3.15), respectively, yields
and
In view of (1.5), (1.6), (3.17) and (3.18), we pose the following conjecture.
Conjecture 3.1
The coefficients \(a_{j}\) and \(b_{j}\) (\(1\leq j\leq k\)) in (3.15) satisfy the following relation:
4 Inequalities for the gamma function
Formulas (3.17) and (3.18) motivate us to establish the following theorem.
Theorem 4.1
The following inequalities hold:
where
and
The left-hand side inequality holds for \(x\geq3\), while the right-hand side inequality is valid for \(x\geq2\).
Proof
It suffices to show that
where
and
Differentiating \(F(x)\) and applying the second inequality in (2.6) yield
where
and
Hence, \(F'(x)<0\) for \(x\geq3\), and we have
Differentiating \(G(x)\) and applying the first inequality in (2.6) yield
where
and
Hence, \(G'(x)>0\) for \(x\geq2\), and we have
The proof is complete. □
Remark 4.1
Following the same method as the one used in the proof of Theorem 4.1, we can prove the double inequality
for \(x\geq2\). We here omit it. Some computer experiments indicate that inequalities (4.1) and (4.4) are valid for \(x\geq1\).
In view of (4.1) and (4.4), we pose the following conjecture.
Conjecture 4.1
If k is odd, then for \(x\geq1\),
where the coefficients \(a_{j}\) and \(b_{j}\) (\(1\leq j \leq k\)) are determined in (3.16). If k is even, then inequality (4.5) is reversed.
5 Comparison
In 2011, Mortici [47] showed by numerical computations that his formula
is much stronger than other known formulas such as:
In 2012, Mahmoud et al. [31] showed numerically that their formula
has a superiority over Mortici’s formula (5.1). Here \(\zeta(s,a)\) denotes the Hurwitz (or generalized) zeta function defined by
\(Z_{0}^{-}\) being the set of nonpositive integers.
From (3.17) and (3.18), we obtain
and
We here offer some numerical computations (see Table 1) to show the superiority of our sequences \((U_{n})_{n\geq1}\) and \((V_{n})_{n\geq1}\) over the sequence \((\lambda_{n})_{n\geq1}\).
References
Abramowitz, M, Stegun, IA (eds.): Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables. Applied Mathematics Series, vol. 55, 9th printing. National Bureau of Standards, Washington (1972)
Copson, ET: Asymptotic Expansions. Cambridge University Press, Cambridge (1965)
Alzer, H: On Ramanujan’s double inequality for the gamma function. Bull. Lond. Math. Soc. 35, 601-607 (2003)
Batir, N: Sharp inequalities for factorial n. Proyecciones 27, 97-102 (2008)
Batir, N: Very accurate approximations for the factorial function. J. Math. Inequal. 4, 335-344 (2010)
Batir, N: Improving Stirling’s formula. Math. Commun. 16, 105-114 (2011)
Burić, T, Elezović, N: New asymptotic expansions of the gamma function and improvements of Stirling’s type formulas. J. Comput. Anal. Appl. 13, 785-795 (2011)
Burnside, W: A rapidly convergent series for \(\log N!\). Messenger Math. 46, 157-159 (1917)
Chen, CP: Unified treatment of several asymptotic formulas for the gamma function. Numer. Algorithms 64, 311-319 (2013)
Chen, CP: Inequalities and asymptotic expansions associated with the Ramanujan and Nemes formulas for the gamma function. Appl. Math. Comput. 261, 337-350 (2015)
Chen, CP: Asymptotic expansions of the gamma function related to Windschitl’s formula. Appl. Math. Comput. 245, 174-180 (2014)
Chen, CP, Batir, N: Some inequalities and monotonicity properties associated with the gamma and psi functions. Appl. Math. Comput. 218, 8217-8225 (2012)
Chen, CP, Lin, L: Remarks on asymptotic expansions for the gamma function. Appl. Math. Lett. 25, 2322-2326 (2012)
Chen, CP, Liu, JY: Complete monotonicity properties and asymptotic expansions of the logarithm of the gamma function. Math. Inequal. Appl. 18, 379-388 (2015)
Chen, CP, Liu, JY: Inequalities and asymptotic expansions for the gamma function. J. Number Theory 149, 313-326 (2015)
Chen, CP, Paris, RB: Inequalities, asymptotic expansions and completely monotonic functions related to the gamma function. Appl. Math. Comput. 250, 514-529 (2015)
Chen, CP: A sharp version of Ramanujan’s inequality for the factorial function. Ramanujan J. 39, 149-154 (2016)
Chen, CP: Monotonicity properties, inequalities and asymptotic expansions associated with the gamma function. Appl. Math. Comput. 283, 385-396 (2016)
Gosper, RW: Decision procedure for indefinite hypergeometric summation. Proc. Natl. Acad. Sci. USA 75, 40-42 (1978)
Hirschhorn, MD: A new version of Stirling’s formula. Math. Gaz. 90, 286-291 (2006)
Hirschhorn, MD, Villarino, MB: A refinement of Ramanujan’s factorial approximation. Ramanujan J. 34, 73-81 (2014)
Karatsuba, EA: On the asymptotic representation of the Euler gamma function by Ramanujan. J. Comput. Appl. Math. 135, 225-240 (2001)
Lin, L: On Stirling’s formula remainder. Appl. Math. Comput. 247, 494-500 (2014)
Lin, L, Chen, CP: Asymptotic formulas for the gamma function by Gosper. J. Math. Inequal. 9, 541-551 (2015)
Lu, D: A generated approximation related to Burnside’s formula. J. Number Theory 136, 414-422 (2014)
Lu, D, Feng, J, Ma, C: A general asymptotic formula of the gamma function based on the Burnside’s formula. J. Number Theory 145, 317-328 (2014)
Lu, D, Song, L, Ma, C: A generated approximation of the gamma function related to Windschitl’s formula. J. Number Theory 140, 215-225 (2014)
Lu, D, Song, L, Ma, C: Some new asymptotic approximations of the gamma function based on Nemes’ formula, Ramanujan’s formula and Burnside’s formula. Appl. Math. Comput. 253, 1-7 (2015)
Lu, D, Wang, X: A generated approximation related to Gosper’s formula and Ramanujan’s formula. J. Math. Anal. Appl. 406, 287-292 (2013)
Lu, D, Wang, X: A new asymptotic expansion and some inequalities for the gamma function. J. Number Theory 140, 314-323 (2014)
Mahmoud, M, Alghamdi, MA, Agarwal, RP: New upper bounds of n!. J. Inequal. Appl. 2012, 27 (2012). http://link.springer.com/article/10.1186/1029-242X-2012-27
Mortici, C: An ultimate extremely accurate formula for approximation of the factorial function. Arch. Math. (Basel) 93, 37-45 (2009)
Mortici, C: Sharp inequalities related to Gosper’s formula. C. R. Math. Acad. Sci. Paris 348, 137-140 (2010)
Mortici, C: Product approximations via asymptotic integration. Am. Math. Mon. 117, 434-441 (2010)
Mortici, C: New improvements of the Stirling formula. Appl. Math. Comput. 217, 699-704 (2010)
Mortici, C: The asymptotic series of the generalized Stirling formula. Comput. Math. Appl. 60, 786-791 (2010)
Mortici, C: Asymptotic expansions of the generalized Stirling approximations. Math. Comput. Model. 52, 1867-1868 (2010)
Mortici, C: A class of integral approximations for the factorial function. Comput. Math. Appl. 59, 2053-2058 (2010)
Mortici, C: Ramanujan formula for the generalized Stirling approximation. Appl. Math. Comput. 217, 2579-2585 (2010)
Mortici, C: Best estimates of the generalized Stirling formula. Appl. Math. Comput. 215, 4044-4048 (2010)
Mortici, C: On the gamma function approximation by Burnside. Appl. Math. E-Notes 11, 274-277 (2011)
Mortici, C: On Gospers formula for the Gamma function. J. Math. Inequal. 5, 611-614 (2011)
Mortici, C: Improved asymptotic formulas for the gamma function. Comput. Math. Appl. 61, 3364-3369 (2011)
Mortici, C: A new fast asymptotic series for the gamma function. Ramanujan J. 38, 549-559 (2015)
Mortici, C: An improvement of the Ramanujan formula for approximation of the Euler gamma function. Carpath. J. Math. 28, 301-304 (2012)
Mortici, C: Ramanujan’s estimate for the gamma function via monotonicity arguments. Ramanujan J. 25, 149-154 (2011)
Mortici, C: A new Stirling series as continued fraction. Numer. Algorithms 56, 17-26 (2011)
Mortici, C: On Ramanujan’s large argument formula for the gamma function. Ramanujan J. 26, 185-192 (2011)
Mortici, C: A continued fraction approximation of the gamma function. J. Math. Anal. Appl. 402, 405-410 (2013)
Mortici, C: Very accurate estimates of the polygamma functions. Asymptot. Anal. 68, 125-134 (2010)
Mortici, C: A new Stirling series as continued fraction. Numer. Algorithms 56, 17-26 (2011)
Nemes, G: Asymptotic expansion for \(\log n!\) in terms of the reciprocal of a triangular number. Acta Math. Hung. 129, 254-262 (2010)
Nemes, G: More accurate approximations for the gamma function. Thai J. Math. 9, 21-28 (2011)
Nemes, G: New asymptotic expansion for the Gamma function. Arch. Math. (Basel) 95, 161-169 (2010)
Alzer, H: On some inequalities for the gamma and psi functions. Math. Comput. 66, 373-389 (1997)
Koumandos, S: Remarks on some completely monotonic functions. J. Math. Anal. Appl. 324, 1458-1461 (2006)
Koumandos, S, Pedersen, HL: Completely monotonic functions of positive order and asymptotic expansions of the logarithm of Barnes double gamma function and Euler’s gamma function. J. Math. Anal. Appl. 355, 33-40 (2009)
Bercu, G: Padé approximant related to remarkable inequalities involving trigonometric functions. J. Inequal. Appl. 2016, 99 (2016). http://www.doc88.com/p-0037658479714.html
Bercu, G: The natural approach of trigonometric inequalities-Pade approximant. J. Math. Inequal. 11, 181-191 (2017)
Bercu, G, Wu, S: Refinements of certain hyperbolic inequalities via the Padé approximation method. J. Nonlinear Sci. Appl. 9, 5011-5020 (2016)
Brezinski, C, Redivo-Zaglia, M: New representations of Padé, Padé-type, and partial Padé approximants. J. Comput. Appl. Math. 284, 69-77 (2015)
Ramanujan, S: The Lost Notebook and Other Unpublished Papers. Springer, Berlin (1988)
Acknowledgements
The authors thank the referees for helpful comments.
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors read and approved the final manuscript.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Li, X., Chen, CP. Padé approximant related to asymptotics for the gamma function. J Inequal Appl 2017, 53 (2017). https://doi.org/10.1186/s13660-017-1315-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s13660-017-1315-1