Die Fakultät (manchmal, besonders in Österreich, auch Faktorielle genannt) ist in der Mathematik eine Funktion , die einer natürlichen Zahl das Produkt aller natürlichen Zahlen (ohne Null) kleiner und gleich dieser Zahl zuordnet. Sie wird durch ein dem Argument nachgestelltes Ausrufezeichen („!“) abgekürzt. Die Fakultät ist die für die natürlichen Zahlen definierte Form der Gaußschen Pifunktion und ordnet alle natürlichzahligen Abszissenwerte natürlichzahligen Ordinatenwerten zu. Die Notation für die Fakultät mit dem Ausrufezeichen wurde erstmals im Jahre 1808 von dem elsässischen Mathematiker Christian Kramp (1760–1826) verwendet, der um 1798 auch die Bezeichnung faculté „Fähigkeit“ dafür einführte.
n
{\displaystyle n}
n
!
{\displaystyle n!}
0
1
1
1
2
2
3
6
4
24
5
120
6
720
7
5.040
8
40.320
9
362.880
10
3.628.800
20
2,432… · 1018 50
3,041… · 1064 100
9,332… · 10157
Für alle natürlichen Zahlen
n
{\displaystyle n}
ist
n
!
=
1
⋅
2
⋅
3
⋯
n
=
∏
k
=
1
n
k
{\displaystyle n!=1\cdot 2\cdot 3\dotsm n=\prod _{k=1}^{n}k}
als das Produkt der natürlichen Zahlen von 1 bis
n
{\displaystyle n}
definiert. Da das leere Produkt stets 1 ist, gilt
0
!
=
1
{\displaystyle 0!=1}
Die Fakultät lässt sich auch rekursiv definieren:
n
!
=
{
1
,
n
=
0
n
⋅
(
n
−
1
)
!
,
n
>
0
{\displaystyle n!={\begin{cases}1,&n=0\\n\cdot (n-1)!,&n>0\end{cases}}}
Fakultäten im ursprünglichen Sinne sind für negative oder nichtganze Zahlen nicht definiert. Es gibt aber eine Erweiterung der Fakultät auf solche Argumente. Sie wird als Gaußsche Pifunktion bezeichnet und ist für alle reellen Zahlen mit Ausnahme der negativen ganzen Zahlen definiert.
Auf
R
∖
Z
<
0
{\displaystyle \mathbb {R} \setminus \mathbb {Z} _{<0}}
kann die Gaußsche Pifunktion beziehungsweise Fakultätsfunktion als Eulersche Gammafunktion der Nachfolgerfunktion definiert werden:
x
!
=
Π
(
x
)
=
Γ
(
x
+
1
)
=
exp
(
−
γ
x
)
∏
n
=
1
∞
[
(
1
+
x
n
)
−
1
exp
(
x
n
)
]
{\displaystyle x!=\Pi (x)=\Gamma (x+1)=\exp(-\,\gamma \,x)\prod _{n=1}^{\infty }{\bigl [}{\bigl (}1+{\frac {x}{n}}{\bigr )}^{-1}\exp {\bigl (}{\frac {x}{n}}{\bigr )}{\bigr ]}}
Das auf der rechten Seite der Gleichungskette gezeigte Produkt wird Weierstraß-Produkt genannt.
Mit
γ
{\displaystyle \gamma }
wird die Euler-Mascheroni-Konstante bezeichnet.
Fakultätswerte von ganzen Zahlen Bearbeiten
0
!
=
1
=
1
1
!
=
1
=
1
2
!
=
1
⋅
2
=
2
3
!
=
1
⋅
2
⋅
3
=
6
4
!
=
1
⋅
2
⋅
3
⋅
4
=
24
5
!
=
1
⋅
2
⋅
3
⋅
4
⋅
5
=
120
⋮
{\displaystyle {\begin{array}{rll}0!&=1&=1\\1!&=1&=1\\2!&=1\cdot 2&=2\\3!&=1\cdot 2\cdot 3&=6\\4!&=1\cdot 2\cdot 3\cdot 4&=24\\5!&=1\cdot 2\cdot 3\cdot 4\cdot 5&=120\\&\;\vdots \end{array}}}
Die Werte der Fakultäten bilden die Folge A000142 in OEIS .
Explizite Fakultätswerte von 0 bis 20
0!
1
1!
1
2!
2
3!
6
4!
24
5!
120
6!
720
7!
5.040
8!
40.320
9!
362.880
10!
3.628.800
11!
39.916.800
12!
479.001.600
13!
6.227.020.800
14!
87.178.291.200
15!
1.307.674.368.000
16!
20.922.789.888.000
17!
355.687.428.096.000
18!
6.402.373.705.728.000
19!
121.645.100.408.832.000
20!
2.432.902.008.176.640.000
Fakultätswerte von Brüchen Bearbeiten
Aus Übersichtlichkeitsgründen werden die Fakultäten der Brüche hier mit der Gaußschen Pifunktion dargestellt und mit Hilfe der Zentralbinomialkoeffizienten (CBC) sowie mit Hilfe des vollständigen elliptischen Integrals erster Art ausgedrückt. Wie oben erwähnt, ist die kontinuierliche Fakultätsfunktion gleich der Gaußschen Pifunktion:
x
!
=
Π
(
x
)
{\displaystyle x!=\Pi (x)}
Für den Zentralbinomialkoeffizienten gilt:
CBC
(
x
)
=
Π
(
2
x
)
Π
(
x
)
−
2
{\displaystyle \operatorname {CBC} (x)=\Pi (2x)\Pi (x)^{-2}}
Fakultät
CBC-Ausdruck
K-Ausdruck
Π
(
1
2
)
{\displaystyle \Pi ({\tfrac {1}{2}})}
CBC
(
1
2
)
−
1
/
2
{\displaystyle \operatorname {CBC} ({\tfrac {1}{2}})^{-1/2}}
1
2
π
1
/
2
{\displaystyle {\tfrac {1}{2}}\pi ^{1/2}}
Π
(
1
3
)
{\displaystyle \Pi ({\tfrac {1}{3}})}
(
4
3
)
1
/
3
CBC
(
1
3
)
−
2
/
3
CBC
(
2
3
)
−
1
/
3
{\displaystyle ({\tfrac {4}{3}})^{1/3}\operatorname {CBC} ({\tfrac {1}{3}})^{-2/3}\operatorname {CBC} ({\tfrac {2}{3}})^{-1/3}}
2
7
/
9
3
−
13
/
12
π
1
/
3
K
[
sin
(
1
12
π
)
]
1
/
3
{\displaystyle 2^{7/9}3^{-13/12}{\pi }^{1/3}{K{\bigl [}\sin({\tfrac {1}{12}}\pi ){\bigr ]}}^{1/3}}
Π
(
2
3
)
{\displaystyle \Pi ({\tfrac {2}{3}})}
(
4
3
)
2
/
3
CBC
(
1
3
)
−
1
/
3
CBC
(
2
3
)
−
2
/
3
{\displaystyle ({\tfrac {4}{3}})^{2/3}\operatorname {CBC} ({\tfrac {1}{3}})^{-1/3}\operatorname {CBC} ({\tfrac {2}{3}})^{-2/3}}
2
11
/
9
3
−
17
/
12
π
2
/
3
K
[
sin
(
1
12
π
)
]
−
1
/
3
{\displaystyle 2^{11/9}3^{-17/12}{\pi }^{2/3}{K{\bigl [}\sin({\tfrac {1}{12}}\pi ){\bigr ]}}^{-1/3}}
Π
(
1
4
)
{\displaystyle \Pi ({\tfrac {1}{4}})}
CBC
(
1
4
)
−
1
/
2
CBC
(
1
2
)
−
1
/
4
{\displaystyle \operatorname {CBC} ({\tfrac {1}{4}})^{-1/2}\operatorname {CBC} ({\tfrac {1}{2}})^{-1/4}}
1
2
π
1
/
4
K
(
1
2
2
)
1
/
2
{\displaystyle {\tfrac {1}{2}}\pi ^{1/4}K({\tfrac {1}{2}}{\sqrt {2}})^{1/2}}
Π
(
3
4
)
{\displaystyle \Pi ({\tfrac {3}{4}})}
(
3
2
)
1
/
2
CBC
(
3
4
)
−
1
/
2
CBC
(
1
2
)
−
1
/
4
{\displaystyle ({\tfrac {3}{2}})^{1/2}\operatorname {CBC} ({\tfrac {3}{4}})^{-1/2}\operatorname {CBC} ({\tfrac {1}{2}})^{-1/4}}
2
−
5
/
2
3
π
3
/
4
K
(
1
2
2
)
−
1
/
2
{\displaystyle 2^{-5/2}3\,\pi ^{3/4}K({\tfrac {1}{2}}{\sqrt {2}})^{-1/2}}
Das sind die Werte der Zentralbinomialkoeffizienten:
CBC
(
1
2
)
=
4
π
{\displaystyle \operatorname {CBC} {\bigl (}{\frac {1}{2}}{\bigr )}={\frac {4}{\pi }}}
CBC
(
1
3
)
=
1
2
4
3
27
4
K
[
sin
(
1
12
π
)
]
−
1
{\displaystyle \operatorname {CBC} {\bigl (}{\frac {1}{3}}{\bigr )}={\frac {1}{2}}{\sqrt[{3}]{4}}\,{\sqrt[{4}]{27}}\,K{\bigl [}\sin {\bigl (}{\frac {1}{12}}\pi {\bigr )}{\bigr ]}^{-1}}
CBC
(
2
3
)
=
1
π
2
3
27
4
K
[
sin
(
1
12
π
)
]
{\displaystyle \operatorname {CBC} {\bigl (}{\frac {2}{3}}{\bigr )}={\frac {1}{\pi }}{\sqrt[{3}]{2}}\,{\sqrt[{4}]{27}}\,K{\bigl [}\sin {\bigl (}{\frac {1}{12}}\pi {\bigr )}{\bigr ]}}
CBC
(
1
4
)
=
2
K
(
1
2
2
)
−
1
=
2
2
ϖ
{\displaystyle \operatorname {CBC} {\bigl (}{\frac {1}{4}}{\bigr )}=2\,K{\bigl (}{\frac {1}{2}}{\sqrt {2}}{\bigr )}^{-1}={\frac {2\,{\sqrt {2}}}{\varpi }}}
CBC
(
3
4
)
=
8
3
π
K
(
1
2
2
)
=
4
2
ϖ
3
π
{\displaystyle \operatorname {CBC} {\bigl (}{\frac {3}{4}}{\bigr )}={\frac {8}{3\,\pi }}K{\bigl (}{\frac {1}{2}}{\sqrt {2}}{\bigr )}={\frac {4\,{\sqrt {2}}\,\varpi }{3\,\pi }}}
Mit dem Kürzel
ϖ
{\displaystyle \varpi }
wird die Lemniskatische Konstante ausgedrückt.
Dabei hat das vollständige elliptische Integral erster Art diese Definition:
K
(
ε
)
=
∫
0
π
/
2
1
1
−
ε
2
sin
(
φ
)
2
d
φ
{\displaystyle K(\varepsilon )=\int _{0}^{\pi /2}{\frac {1}{\sqrt {1-\varepsilon ^{2}\sin(\varphi )^{2}}}}\,\mathrm {d} \varphi }
K
(
ε
)
=
∫
0
1
2
(
x
2
+
1
)
2
−
4
ε
2
x
2
d
x
{\displaystyle K(\varepsilon )=\int _{0}^{1}{\frac {2}{\sqrt {(x^{2}+1)^{2}-4\,\varepsilon ^{2}x^{2}}}}\,\mathrm {d} x}
Diese beiden nun genannten Definitionsformeln für das elliptische K-Integral stimmen miteinander überein.
Die gezeigten Werte des Zentralbinomialkoeffizienten können sehr leicht durch diese Formeln[1] erzeugt werden:
CBC
(
n
)
=
(
2
n
n
)
=
2
2
n
+
1
π
∫
0
∞
d
x
(
x
2
+
1
)
n
+
1
{\displaystyle \operatorname {CBC} (n)={\binom {2n}{n}}={\frac {2^{2n+1}}{\pi }}\int \limits _{0}^{\infty }{\frac {\mathrm {d} x}{(x^{2}+1)^{n+1}}}}
1
CBC
(
n
)
=
∫
0
1
n
x
n
−
1
(
x
+
1
)
2
n
d
x
{\displaystyle {\frac {1}{\operatorname {CBC} (n)}}=\int _{0}^{1}{\frac {n\,x^{n-1}}{(x+1)^{2n}}}\,\mathrm {d} x}
Generell gilt dann folgende Formel für alle
m
∈
N
{\displaystyle m\in \mathbb {N} }
:
1
CBC
(
1
/
m
)
=
Π
(
1
/
m
)
2
Π
(
2
/
m
)
=
∫
0
1
1
(
x
m
+
1
)
2
/
m
d
x
=
∫
0
1
1
2
2
/
m
1
−
x
m
d
x
{\displaystyle {\frac {1}{\operatorname {CBC} (1/m)}}={\frac {\Pi (1/m)^{2}}{\Pi (2/m)}}=\int _{0}^{1}{\frac {1}{(x^{m}+1)^{2/m}}}\,\mathrm {d} x=\int _{0}^{1}{\frac {1}{2^{2/m}{\sqrt {1-x^{m}}}}}\,\mathrm {d} x}
Im letzten Schritt wird auf folgende Weise substituiert:
x
↦
x
(
1
+
1
−
x
m
)
2
/
m
{\displaystyle x\mapsto {\frac {x}{(1+{\sqrt {1-x^{m}}})^{2/m}}}}
Das zuletzt genannte Integral ermöglicht die Ermittlung der CBC-Werte von Brüchen mit Hilfe einfacher elliptischer Stammfunktionen.
Die Pifunktionswerte der Kehrwerte der Mersenne-Zahlen ergeben sich stets durch Multiplikation von Potenzen der Zentralbinomialkoeffizienten aus den Zweierpotenz-Vielfachen der genannten Kehrwerte:
Fakultät
CBC-Ausdruck
Π
(
1
7
)
{\displaystyle \Pi ({\tfrac {1}{7}})}
(
8
7
)
1
/
7
CBC
(
1
7
)
−
4
/
7
CBC
(
2
7
)
−
2
/
7
CBC
(
4
7
)
−
1
/
7
{\displaystyle ({\tfrac {8}{7}})^{1/7}\operatorname {CBC} ({\tfrac {1}{7}})^{-4/7}\operatorname {CBC} ({\tfrac {2}{7}})^{-2/7}\operatorname {CBC} ({\tfrac {4}{7}})^{-1/7}}
Π
(
1
15
)
{\displaystyle \Pi ({\tfrac {1}{15}})}
(
16
15
)
1
/
15
CBC
(
1
15
)
−
8
/
15
CBC
(
2
15
)
−
4
/
15
CBC
(
4
15
)
−
2
/
15
CBC
(
8
15
)
−
1
/
15
{\displaystyle ({\tfrac {16}{15}})^{1/15}\operatorname {CBC} ({\tfrac {1}{15}})^{-8/15}\operatorname {CBC} ({\tfrac {2}{15}})^{-4/15}\operatorname {CBC} ({\tfrac {4}{15}})^{-2/15}\operatorname {CBC} ({\tfrac {8}{15}})^{-1/15}}
Π
(
1
31
)
{\displaystyle \Pi ({\tfrac {1}{31}})}
(
32
31
)
1
/
31
CBC
(
1
31
)
−
16
/
31
CBC
(
2
31
)
−
8
/
31
CBC
(
4
31
)
−
4
/
31
CBC
(
8
31
)
−
2
/
31
CBC
(
16
31
)
−
1
/
31
{\displaystyle ({\tfrac {32}{31}})^{1/31}\operatorname {CBC} ({\tfrac {1}{31}})^{-16/31}\operatorname {CBC} ({\tfrac {2}{31}})^{-8/31}\operatorname {CBC} ({\tfrac {4}{31}})^{-4/31}\operatorname {CBC} ({\tfrac {8}{31}})^{-2/31}\operatorname {CBC} ({\tfrac {16}{31}})^{-1/31}}
Π
(
1
63
)
{\displaystyle \Pi ({\tfrac {1}{63}})}
(
64
63
)
1
/
63
CBC
(
1
63
)
−
32
/
63
CBC
(
2
63
)
−
16
/
63
CBC
(
4
63
)
−
8
/
63
CBC
(
8
63
)
−
4
/
63
CBC
(
16
63
)
−
2
/
63
CBC
(
32
63
)
−
1
/
63
{\displaystyle ({\tfrac {64}{63}})^{1/63}\operatorname {CBC} ({\tfrac {1}{63}})^{-32/63}\operatorname {CBC} ({\tfrac {2}{63}})^{-16/63}\operatorname {CBC} ({\tfrac {4}{63}})^{-8/63}\operatorname {CBC} ({\tfrac {8}{63}})^{-4/63}\operatorname {CBC} ({\tfrac {16}{63}})^{-2/63}\operatorname {CBC} ({\tfrac {32}{63}})^{-1/63}}
Das grundlegendste Theorem über die Fakultätsfunktion ist das folgende Theorem:
x
!
=
Π
(
x
)
=
x
Π
(
x
−
1
)
{\displaystyle x!=\Pi (x)=x\,\Pi (x-1)}
Eulerscher Ergänzungssatz Bearbeiten
Im Jahre 1749 hat der schweizerische Mathematiker Leonhard Euler einen Ergänzungssatz[2] [3] entdeckt, welcher nach ihm benannt wurde. Im nun Folgenden wird der Eulersche Ergänzungssatz mittels Gaußscher Pifunktion dargestellt:
Π
(
x
)
Π
(
1
−
x
)
=
π
x
(
1
−
x
)
csc
(
π
x
)
{\displaystyle \Pi (x)\,\Pi (1-x)=\pi \,x(1-x)\csc(\pi \,x)}
Legendresche Verdopplungsformel Bearbeiten
Im Jahre 1809 hat der französische Mathematiker Adrien Marie Legendre die Verdopplungsformel für die Fakultät[4] entdeckt, welcher mittels Gammafunktionsausdrücken in der Sammlung Mémoires de l'Institut des Sciences et Arts aus dem Institut de France verewigt ist. Auch diese Identität wird hier mittels Gaußscher Pifunktion dargestellt:
Π
(
x
)
Π
(
x
+
1
2
)
=
π
1
/
2
2
−
2
x
−
1
(
2
x
+
1
)
Π
(
2
x
)
{\displaystyle \Pi (x)\,\Pi (x+{\tfrac {1}{2}})=\pi ^{1/2}\,2^{-2x-1}(2x+1)\,\Pi (2x)}
Herleitungen über die Fakultätsfunktion Bearbeiten
Herleitung des Eulerschen Integrals Bearbeiten
Gegeben ist die diskrete und ebenso ursprünglichste Definition der Fakultätsfunktion für alle natürlichen Zahlen
x
∈
N
{\displaystyle x\in \mathbb {N} }
nach dem oben genannten Muster:
x
!
=
∏
n
=
1
x
n
{\displaystyle x!=\prod _{n=1}^{x}n}
Diese diskret definierte Funktionsdefinition erfüllt folgendes Induktionskriterium aus folgenden zwei verknüpften Formeln:
x
!
=
x
(
x
−
1
)
!
{\displaystyle x!=x\,(x-1)!}
0
!
=
1
{\displaystyle 0!=1}
Das Eulersche Integral zweiter Art oder auch das Eulersche Integral zweiter Gattung definiert die Fakultätsfunktion beziehungsweise Gaußsche Pifunktion für alle Zahlen größer als Minus Eins:
x
!
=
Π
(
x
)
=
∫
0
∞
y
x
exp
(
−
y
)
d
y
{\displaystyle x!=\Pi (x)=\int _{0}^{\infty }y^{x}\exp(-y)\,\mathrm {d} y}
Denn diese Funktion erfüllt die genannte Induktion:
f
(
x
)
=
∫
0
∞
y
x
exp
(
−
y
)
d
y
{\displaystyle f(x)=\int _{0}^{\infty }y^{x}\exp(-y)\,\mathrm {d} y}
f
(
x
)
=
∫
0
∞
y
x
exp
(
−
y
)
d
y
=
∫
0
∞
x
y
x
−
1
exp
(
−
y
)
d
y
−
∫
0
∞
[
x
y
x
−
1
exp
(
−
y
)
−
y
x
exp
(
−
y
)
]
d
y
=
{\displaystyle f(x)=\int _{0}^{\infty }y^{x}\exp(-y)\,\mathrm {d} y=\int _{0}^{\infty }x\,y^{x-1}\exp(-y)\,\mathrm {d} y-\int _{0}^{\infty }{\bigl [}x\,y^{x-1}\exp(-y)-y^{x}\exp(-y){\bigr ]}\,\mathrm {d} y=}
=
∫
0
∞
x
y
x
−
1
exp
(
−
y
)
d
y
−
[
y
x
exp
(
−
y
)
]
y
=
0
y
=
∞
=
∫
0
∞
x
y
x
−
1
exp
(
−
y
)
d
y
=
x
f
(
x
−
1
)
mit allen
x
∈
R
+
{\displaystyle =\int _{0}^{\infty }x\,y^{x-1}\exp(-y)\,\mathrm {d} y-{\biggl [}y^{x}\exp(-y){\biggr ]}_{y=0}^{y=\infty }=\int _{0}^{\infty }x\,y^{x-1}\exp(-y)\,\mathrm {d} y=xf(x-1)\,\,\,\,\,{\text{mit allen}}\,\,x\in \mathbb {R} ^{+}}
f
(
0
)
=
∫
0
∞
y
0
exp
(
−
y
)
d
y
=
∫
0
∞
exp
(
−
y
)
d
y
=
1
{\displaystyle f(0)=\int _{0}^{\infty }y^{0}\exp(-y)\,\mathrm {d} y=\int _{0}^{\infty }\exp(-y)\,\mathrm {d} y=1}
f
(
x
)
=
x
!
=
Π
(
x
)
{\displaystyle f(x)=x!=\Pi (x)}
Aus
f
(
x
)
=
x
f
(
x
−
1
)
{\displaystyle f(x)=xf(x-1)}
und
f
(
0
)
=
1
{\displaystyle f(0)=1}
folgt
f
(
x
)
=
x
!
{\displaystyle f(x)=x!}
direkt.
Herleitung der Stammfunktion von H über Induktion Bearbeiten
Für alle reellen Zahlen
x
∈
R
∖
Z
<
0
{\displaystyle x\in \mathbb {R} \setminus \mathbb {Z} _{<0}}
ist die harmonische Reihenfunktion nach Karl Weierstraß so definiert:
H
(
x
)
=
∑
n
=
1
∞
(
1
n
−
1
n
+
x
)
{\displaystyle \mathrm {H} (x)=\sum _{n=1}^{\infty }{\bigl (}{\frac {1}{n}}-{\frac {1}{n+x}}{\bigr )}}
An der Summenreihe ist das Rekursionskriterium ablesbar:
H
(
x
+
1
)
=
H
(
x
)
+
1
x
+
1
{\displaystyle \mathrm {H} (x+1)=\mathrm {H} (x)+{\frac {1}{x+1}}}
Für alle Zahlen
w
∈
N
{\displaystyle w\in \mathbb {N} }
gilt dann auch diese Summe:
H
(
x
+
w
)
=
H
(
x
)
+
∑
v
=
1
w
1
x
+
v
{\displaystyle \mathrm {H} (x+w)=\mathrm {H} (x)+\sum _{v=1}^{w}{\frac {1}{x+v}}}
Das Integral der harmonischen Reihenfunktion führt direkt zur Definition der Mascheronischen Konstante:
∫
0
1
H
(
x
)
d
x
=
∫
0
1
∑
n
=
1
∞
(
1
n
−
1
n
+
x
)
d
x
=
∑
n
=
1
∞
∫
0
1
(
1
n
−
1
n
+
x
)
d
x
=
{\displaystyle \int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x=\int _{0}^{1}\sum _{n=1}^{\infty }{\bigl (}{\frac {1}{n}}-{\frac {1}{n+x}}{\bigr )}\,\mathrm {d} x=\sum _{n=1}^{\infty }\int _{0}^{1}{\bigl (}{\frac {1}{n}}-{\frac {1}{n+x}}{\bigr )}\,\mathrm {d} x=}
=
∑
n
=
1
∞
[
x
n
−
ln
(
1
+
x
n
)
]
x
=
0
x
=
∞
=
∑
n
=
1
∞
[
1
n
−
ln
(
1
+
1
n
)
]
=
γ
{\displaystyle =\sum _{n=1}^{\infty }{\biggl [}{\frac {x}{n}}-\ln {\biggl (}1+{\frac {x}{n}}{\biggr )}{\biggr ]}_{x=0}^{x=\infty }=\sum _{n=1}^{\infty }{\biggl [}{\frac {1}{n}}-\ln {\biggl (}1+{\frac {1}{n}}{\biggr )}{\biggr ]}=\gamma }
Im Folgenden werden die Integralgrenzen verschoben. So ist dann für alle natürlichen Werte
w
∈
N
{\displaystyle w\in \mathbb {N} }
die folgende Beziehung gültig:
∫
w
w
+
1
H
(
x
)
d
x
=
∫
0
1
H
(
x
+
w
)
d
x
=
∫
0
1
[
H
(
x
)
+
∑
v
=
1
w
1
x
+
v
]
d
x
=
{\displaystyle \int _{w}^{w+1}\mathrm {H} (x)\,\mathrm {d} x=\int _{0}^{1}\mathrm {H} (x+w)\,\mathrm {d} x=\int _{0}^{1}{\biggl [}\mathrm {H} (x)+\sum _{v=1}^{w}{\frac {1}{x+v}}{\biggr ]}\,\mathrm {d} x=}
=
∫
0
1
H
(
x
)
d
x
+
∫
0
1
∑
v
=
1
w
1
x
+
v
d
x
=
∫
0
1
H
(
x
)
d
x
+
∑
v
=
1
w
∫
0
1
1
x
+
v
d
x
=
{\displaystyle =\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\int _{0}^{1}\sum _{v=1}^{w}{\frac {1}{x+v}}\,\mathrm {d} x=\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\sum _{v=1}^{w}\int _{0}^{1}{\frac {1}{x+v}}\,\mathrm {d} x=}
=
∫
0
1
H
(
x
)
d
x
+
∑
v
=
1
w
[
ln
(
1
+
x
v
)
]
x
=
0
x
=
∞
=
∫
0
1
H
(
x
)
d
x
+
∑
v
=
1
w
ln
(
1
+
1
v
)
=
{\displaystyle =\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\sum _{v=1}^{w}{\biggl [}\ln {\bigl (}1+{\frac {x}{v}}{\bigr )}{\biggr ]}_{x=0}^{x=\infty }=\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\sum _{v=1}^{w}\ln {\bigl (}1+{\frac {1}{v}}{\bigr )}=}
=
∫
0
1
H
(
x
)
d
x
+
ln
(
w
+
1
)
=
γ
+
ln
(
w
+
1
)
{\displaystyle =\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\ln(w+1)=\gamma +\ln(w+1)}
Das Resultat dieser Gleichungskette lautet somit wie folgt:
∫
w
w
+
1
H
(
x
)
d
x
=
γ
+
ln
(
w
+
1
)
{\displaystyle \int _{w}^{w+1}\mathrm {H} (x)\,\mathrm {d} x=\gamma +\ln(w+1)}
Aus diesem Resultat folgt durch Induktion diese Überleitung:
∫
0
z
+
1
H
(
x
)
d
x
=
∫
0
1
H
(
x
)
d
x
+
∑
u
=
1
z
∫
u
u
+
1
H
(
x
)
d
x
=
{\displaystyle \int _{0}^{z+1}\mathrm {H} (x)\,\mathrm {d} x=\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\sum _{u=1}^{z}\int _{u}^{u+1}\mathrm {H} (x)\,\mathrm {d} x=}
=
∫
0
1
H
(
x
)
d
x
+
∑
u
=
1
z
[
γ
+
ln
(
u
+
1
)
]
=
γ
+
∑
u
=
1
z
[
γ
+
ln
(
u
+
1
)
]
=
{\displaystyle =\int _{0}^{1}\mathrm {H} (x)\,\mathrm {d} x+\sum _{u=1}^{z}{\bigl [}\gamma +\ln(u+1){\bigr ]}=\gamma +\sum _{u=1}^{z}{\bigl [}\gamma +\ln(u+1){\bigr ]}=}
=
γ
(
u
+
1
)
+
∑
u
=
1
z
ln
(
u
+
1
)
=
γ
(
z
+
1
)
+
ln
[
∏
u
=
1
z
(
u
+
1
)
]
=
{\displaystyle =\gamma \,(u+1)+\sum _{u=1}^{z}\ln(u+1)=\gamma \,(z+1)+\ln {\biggl [}\prod _{u=1}^{z}(u+1){\biggr ]}=}
=
γ
(
z
+
1
)
+
ln
[
(
z
+
1
)
!
]
{\displaystyle =\gamma \,(z+1)+\ln {\bigl [}(z+1)!{\bigr ]}}
Denn die Summe der Logarithmen ist gleich dem Logarithmus vom Produkt.
Direkt daraus entsteht dann die Ursprungsstammfunktion der harmonischen Reihenfunktion:
∫
0
x
H
(
y
)
d
y
=
γ
x
+
ln
(
x
!
)
=
γ
x
+
ln
[
Π
(
x
)
]
{\displaystyle \int _{0}^{x}\mathrm {H} (y)\,\mathrm {d} y=\gamma \,x+\ln(x!)=\gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}}
Die Fakultätsfunktion beziehungsweise Gaußsche Pifunktion wird somit auf folgende Weise abgeleitet:
d
d
x
Π
(
x
)
=
Π
(
x
)
[
H
(
x
)
−
γ
]
{\displaystyle {\frac {\mathrm {d} }{\mathrm {d} x}}\,\Pi (x)=\Pi (x){\bigl [}\mathrm {H} (x)-\gamma {\bigr ]}}
Herleitung der Produktreihe nach Weierstraß Bearbeiten
Nach der vorherigen Herleitung gilt für alle Werte
x
>
−
1
{\displaystyle x>-1}
auch diese Formel:
d
d
x
{
γ
x
+
ln
[
Π
(
x
)
]
}
=
H
(
x
)
{\displaystyle {\frac {\mathrm {d} }{\mathrm {d} x}}{\bigl \{}\gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}{\bigr \}}=\operatorname {H} (x)}
Wie beschrieben hat die harmonischen Reihenfunktion diese Definition:
H
(
x
)
=
∑
n
=
1
∞
(
1
n
−
1
n
+
x
)
{\displaystyle \mathrm {H} (x)=\sum _{n=1}^{\infty }{\bigl (}{\frac {1}{n}}-{\frac {1}{n+x}}{\bigr )}}
Durch Bildung der Ursprungsstammfunktion bezüglich x folgt dann diese Gleichung:
γ
x
+
ln
[
Π
(
x
)
]
=
∑
n
=
1
∞
[
x
n
−
ln
(
1
+
x
n
)
]
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\sum _{n=1}^{\infty }{\biggl [}{\frac {x}{n}}-\ln {\biggl (}1+{\frac {x}{n}}{\biggr )}{\biggr ]}}
Die zweite Potenzregel besagt, dass die Exponentialfunktion aus einer Summe gleich dem Produkt aus den Exponentialfunktionen ist:
exp
{
γ
x
+
ln
[
Π
(
x
)
]
}
=
∏
n
=
1
∞
exp
[
x
n
−
ln
(
1
+
x
n
)
]
{\displaystyle \exp {\bigl \{}\gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}{\bigr \}}=\prod _{n=1}^{\infty }\exp {\biggl [}{\frac {x}{n}}-\ln {\biggl (}1+{\frac {x}{n}}{\biggr )}{\biggr ]}}
Durch weitere Anwendung der zweiten Potenzregel entsteht folgender Ausdruck:
Π
(
x
)
exp
(
γ
x
)
=
∏
n
=
1
∞
(
1
+
x
n
)
−
1
exp
(
x
n
)
{\displaystyle \Pi (x)\exp(\gamma \,x)=\prod _{n=1}^{\infty }{\biggl (}1+{\frac {x}{n}}{\biggr )}^{-1}\exp {\biggl (}{\frac {x}{n}}{\biggr )}}
So kommt direkt die Produktreihe nach Weierstraß hervor, die für alle Zahlen
x
∈
R
∖
Z
<
0
{\displaystyle x\in \mathbb {R} \setminus \mathbb {Z} _{<0}}
Gültigkeit hat:
x
!
=
Π
(
x
)
=
exp
(
−
γ
x
)
∏
n
=
1
∞
(
1
+
x
n
)
−
1
exp
(
x
n
)
{\displaystyle x!=\Pi (x)=\exp(-\,\gamma \,x)\prod _{n=1}^{\infty }{\biggl (}1+{\frac {x}{n}}{\biggr )}^{-1}\exp {\biggl (}{\frac {x}{n}}{\biggr )}}
In der abzählenden Kombinatorik spielen Fakultäten eine wichtige Rolle, weil
n
!
{\displaystyle n!}
die Anzahl der Möglichkeiten ist,
n
{\displaystyle n}
unterscheidbare Gegenstände in einer Reihe anzuordnen. Falls
X
{\displaystyle X}
eine
n
{\displaystyle n}
-elementige Menge ist, so ist
n
!
{\displaystyle n!}
auch die Anzahl der bijektiven Abbildungen
X
→
X
{\displaystyle X\to X}
(die Anzahl der Permutationen ). Dies gilt insbesondere auch für den Fall
n
=
0
{\displaystyle n=0}
, da es genau eine Möglichkeit gibt, die leere Menge auf sich selbst abzubilden.
Beispielsweise gibt es bei einem Autorennen mit sechs Fahrern
6
!
{\displaystyle 6!}
verschiedene Möglichkeiten für die Reihenfolge beim Zieleinlauf, wenn alle Fahrer das Ziel erreichen. Für den ersten Platz kommen alle sechs Fahrer in Frage. Ist der erste Fahrer angekommen, können nur noch fünf Fahrer um den zweiten Platz konkurrieren. Für die Belegung des zweiten Platzes ist es maßgeblich, welcher der sechs Fahrer nicht berücksichtigt werden muss (da er bereits auf Rang 1 platziert ist). Daher muss für jede Belegungsmöglichkeit von Platz 1 gesondert gezählt werden, wie viele Belegungsmöglichkeiten für Platz 2 bestehen. Für die Belegung der Plätze 1 und 2 ergeben sich bei sechs Fahrern daher
6
⋅
5
{\displaystyle 6\cdot 5}
Möglichkeiten. Ist auch der zweite Platz vergeben, kommen für den dritten Platz nur noch vier Fahrer in Frage, woraus sich für die ersten drei Plätze und sechs Fahrer
6
⋅
5
⋅
4
{\displaystyle 6\cdot 5\cdot 4}
Belegungsmöglichkeiten ergeben usw. Letztlich gibt es also
6
!
=
6
⋅
5
⋅
4
⋅
3
⋅
2
⋅
1
=
720
{\displaystyle 6!=6\cdot 5\cdot 4\cdot 3\cdot 2\cdot 1=720}
verschiedene Ranglisten für den Zieleinlauf.
Binomialkoeffizienten Bearbeiten
Ein Begriff, der in der abzählenden Kombinatorik eine ähnlich zentrale Stellung wie die Fakultät einnimmt, ist der Binomialkoeffizient
(
n
k
)
=
n
!
k
!
(
n
−
k
)
!
{\displaystyle {n \choose k}={\frac {n!}{k!\,(n-k)!}}}
.Er gibt die Anzahl der Möglichkeiten an, eine
k
{\displaystyle k}
-elementige Teilmenge aus einer
n
{\displaystyle n}
-elementigen Menge auszuwählen. Umgekehrt gilt
n
!
=
∑
i
=
0
n
−
1
(
−
1
)
i
(
n
n
−
i
)
(
n
−
i
)
n
{\displaystyle n!=\sum _{i=0}^{n-1}(-1)^{i}{\binom {n}{n-i}}(n-i)^{n}}
.Das beliebteste Beispiel ist das Zahlenlotto 6 aus 49 mit insgesamt 13983816 Möglichkeiten, sich sechs verschiedene Kugeln auszusuchen.
(
49
6
)
=
49
!
6
!
(
49
−
6
)
!
=
13
983
816
{\displaystyle {49 \choose 6}={\frac {49!}{6!\,(49-6)!}}=13\,983\,816}
Das bedeutet, dass die Wahrscheinlichkeit, bei dem Lottospiel 6 aus 49 zu gewinnen, nur 1/13983816 und somit weniger als ein Zehnmillionstel beträgt.
Ein weiteres bekanntes Beispiel ist der Sack voller farbiger Murmeln. Die Wahrscheinlichkeit, aus insgesamt
a
{\displaystyle {\color {red}\mathrm {a} }}
roten,
b
{\displaystyle {\color {green}\mathrm {b} }}
grünen und
c
{\displaystyle {\color {blue}\mathrm {c} }}
blauen Murmeln genau
d
{\displaystyle {\color {Red}\mathrm {d} }}
rote,
e
{\displaystyle {\color {Green}\mathrm {e} }}
grüne und
f
{\displaystyle {\color {Blue}\mathrm {f} }}
blaue Murmeln beim Herausholen zu bekommen, wobei man insgesamt
d
+
e
+
f
{\displaystyle \mathrm {d} +\mathrm {e} +\mathrm {f} }
Murmeln herausholen soll, hat folgenden Wert:
P
=
(
a
d
)
(
b
e
)
(
c
f
)
÷
(
a
+
b
+
c
d
+
e
+
f
)
=
a
!
b
!
c
!
(
d
+
e
+
f
)
!
(
a
+
b
+
c
−
d
−
e
−
f
)
!
d
!
e
!
f
!
(
a
−
d
)
!
(
b
−
e
)
!
(
c
−
f
)
!
(
a
+
b
+
c
)
!
{\displaystyle P={\binom {\color {red}a}{\color {Red}d}}{\binom {\color {green}b}{\color {Green}e}}{\binom {\color {blue}c}{\color {Blue}f}}\div {\binom {{\color {red}a}+{\color {green}b}+{\color {blue}c}}{{\color {Red}d}+{\color {Green}e}+{\color {Blue}f}}}={\frac {a!\,b!\,c!\,(d+e+f)!\,(a+b+c-d-e-f)!}{d!\,e!\,f!\,(a-d)!\,(b-e)!\,(c-f)!\,(a+b+c)!}}}
Wenn beispielsweise aus einem Murmelsack mit insgesamt
4
{\displaystyle {\color {red}\mathrm {4} }}
roten,
5
{\displaystyle {\color {green}\mathrm {5} }}
grünen und
6
{\displaystyle {\color {blue}\mathrm {6} }}
blauen Murmeln genau sechs Murmeln blind herausgenommen werden sollen, dann beträgt die Wahrscheinlichkeit, dass sich unter den sechs herausgenommenen Murmeln genau
1
{\displaystyle {\color {Red}\mathrm {1} }}
rote,
2
{\displaystyle {\color {Green}\mathrm {2} }}
grüne und
3
{\displaystyle {\color {Blue}\mathrm {3} }}
blaue Murmeln befinden, exakt
160
/
1001
{\displaystyle 160/1001}
:
P
=
(
4
1
)
(
5
2
)
(
6
3
)
÷
(
4
+
5
+
6
1
+
2
+
3
)
=
4
×
10
×
20
5005
=
160
1001
=
0
,
159840
¯
{\displaystyle P={\binom {\color {red}4}{\color {Red}1}}{\binom {\color {green}5}{\color {Green}2}}{\binom {\color {blue}6}{\color {Blue}3}}\div {\binom {{\color {red}4}+{\color {green}5}+{\color {blue}6}}{{\color {Red}1}+{\color {Green}2}+{\color {Blue}3}}}={\frac {4\times 10\times 20}{5005}}={\frac {160}{1001}}=0{,}{\overline {159840}}}
Das Geburtstagsproblem ist eines der bekanntesten stochastisch kombinatorischen Rätsel über die Fakultät. Dieses Rätsel behandelt die Wahrscheinlichkeit, mit welcher unter einer gegebenen Gruppe von insgesamt n Personen mindestens zwei Personen aus dieser Gruppe zueinander am gleichen Tag Geburtstag haben. Gegeben ist hierbei ein Nicht-Schaltjahr, also ein Jahr aus 365 Tagen. In Abhängigkeit von der Personenzahl n wird diese Wahrscheinlichkeit mit dieser Formel berechnet:
P
(
n
)
=
1
−
365
!
(
365
−
n
)
!
365
n
{\displaystyle P(n)=1-{\frac {365!}{(365-n)!\,365^{n}}}}
Beispielsweise[5] beträgt die Wahrscheinlichkeit, dass unter zehn Personen ein gemeinsamer Geburtstag auftaucht mehr als zehn Prozent, und die Wahrscheinlichkeit, dass unter fünfzehn Personen ein gemeinsamer Geburtstag auftaucht, mehr als fünfundzwanzig Prozent:
P
(
10
)
=
1
−
365
!
355
!
365
10
≈
0,116
9481777110776
{\displaystyle P(10)=1-{\frac {365!}{355!\,365^{10}}}\approx 0{,}1169481777110776}
P
(
15
)
=
1
−
365
!
350
!
365
15
≈
0,252
9013197636863
{\displaystyle P(15)=1-{\frac {365!}{350!\,365^{15}}}\approx 0{,}2529013197636863}
Eine prominente Stelle, an der Fakultäten vorkommen, sind die Taylorreihen vieler Funktionen wie zum Beispiel der Sinusfunktion und der Exponentialfunktion .
Die eulersche Zahl
e
{\displaystyle \mathrm {e} }
lässt sich als Summe der Kehrwerte der Fakultäten definieren:
e
=
∑
k
=
0
∞
1
k
!
=
1
0
!
+
1
1
!
+
1
2
!
+
1
3
!
+
1
4
!
+
⋯
{\displaystyle \mathrm {e} =\sum _{k=0}^{\infty }{\frac {1}{k!}}={\frac {1}{0!}}+{\frac {1}{1!}}+{\frac {1}{2!}}+{\frac {1}{3!}}+{\frac {1}{4!}}+\dotsb }
Numerische Berechnung und Näherung Bearbeiten
Rekursive und iterative Berechnung Bearbeiten
Der numerische Wert für
n
!
{\displaystyle n!}
kann gut rekursiv oder iterativ berechnet werden, falls
n
{\displaystyle n}
nicht zu groß ist.
Die größte Fakultät, die von den meisten handelsüblichen Taschenrechnern berechnet werden kann, ist
69
!
≈
1
,
7
⋅
10
98
,
{\displaystyle 69!\approx 1{,}7\cdot 10^{98},}
da
70
!
≈
1
,
2
⋅
10
100
{\displaystyle 70!\approx 1{,}2\cdot 10^{100}}
außerhalb des üblicherweise verfügbaren Zahlenbereiches liegt.
Die größte als Gleitkommazahl im Format double precision des IEEE-754-Standards darstellbare Fakultät ist
170
!
≈
7
,
3
⋅
10
306
{\displaystyle 170!\approx 7{,}3\cdot 10^{306}}
.
Mit Bibliotheken für sehr große Ganzzahlen (keine Limitierung auf 32, 64 oder z. B. 512 Bit) benötigt zum Beispiel ein Intel Pentium 4 für die Berechnung von 10000! nur wenige Sekunden. Die Zahl hat 35660 Stellen in der Dezimaldarstellung , wobei die letzten 2499 Stellen nur aus der Ziffer Null bestehen.
# Syntax: Python 3.7
n = int ( input ( 'Fakultät von n = ' ))
f = 1
for i in range ( 1 , n + 1 ):
f *= i
print ( f ' { n } ! = { f } ' )
Rekursive Lösung
def fak ( n : int ) -> int :
return 1 if n <= 1 else n * fak ( n - 1 )
Näherung mit der Stirling-Formel Bearbeiten
Wenn
n
{\displaystyle n}
groß ist, bekommt man eine gute Näherung für
n
!
{\displaystyle n!}
mit Hilfe der Stirling-Formel :
n
!
∼
2
π
n
(
n
e
)
n
{\displaystyle n!\sim {\sqrt {2\pi n}}\left({\frac {n}{\mathrm {e} }}\right)^{n}}
Dabei bedeutet
∼
{\displaystyle \sim }
, dass der Quotient aus linker und rechter Seite für
n
→
∞
{\displaystyle n\to \infty }
gegen
1
{\displaystyle 1}
konvergiert.
Durch Approximation (statt Abschneiden) der Stirling-Reihe gelang Bill Gosper eine noch bessere Näherung:[6]
n
!
∼
(
2
n
+
1
/
3
)
π
(
n
e
)
n
{\displaystyle n!\sim {\sqrt {(2n+1/3)\pi }}\left({\frac {n}{\mathrm {e} }}\right)^{n}}
Fakultät-ähnliche Funktionen Bearbeiten
Es gibt eine Reihe weiterer Folgen und Funktionen, die in ihrer Definition oder ihren Eigenschaften ähnlich aussehen wie die Fakultät:
Die Gammafunktion
Γ
(
z
)
{\displaystyle \Gamma (z)}
verallgemeinert die Fakultät um ihren Definitionsbereich von den natürlichen bis hin zu den komplexen Zahlen :[7]
z
!
=
Γ
(
z
+
1
)
,
z
∈
C
{\displaystyle z!=\Gamma (z+1),\qquad z\in \mathbb {C} }
,
ℜ
(
z
)
>
0
{\displaystyle \Re {(z)}>0}
Γ
(
z
)
=
∫
0
∞
t
z
−
1
e
−
t
d
t
=
∫
0
1
(
−
log
t
)
z
−
1
d
t
{\displaystyle \Gamma (z)=\int \limits _{0}^{\infty }t^{z-1}\mathrm {e} ^{-t}\mathrm {d} t=\int _{0}^{1}{(-\log {t})^{z-1}\mathrm {d} t}}
Für
z
∈
C
∖
Z
≤
0
{\displaystyle z\in \mathbb {C} \setminus \mathbb {Z_{\leq 0}} }
kann die Gammafunktion folgendermaßen erweitert werden:[8]
Γ
(
z
)
=
∑
n
=
0
∞
(
−
1
)
n
n
!
(
n
+
z
)
+
∫
1
∞
t
z
−
1
e
−
t
d
t
{\displaystyle \Gamma (z)=\sum _{n=0}^{\infty }{\frac {(-1)^{n}}{n!(n+z)}}+\int _{1}^{\infty }t^{z-1}e^{-t}\mathrm {d} t}
Eine kombinatorische Verallgemeinerung stellen die steigenden und fallenden Faktoriellen
(
n
)
k
{\displaystyle (n)_{k}}
und
(
n
)
k
{\displaystyle (n)^{k}}
dar, denn
(
n
)
n
=
(
1
)
n
=
n
!
{\displaystyle (n)_{n}=(1)^{n}=n!}
.
Primorial (Primfakultät) Bearbeiten
n
n#
n
n#
1
1
5
30
2
2
6
30
3
6
7
210
4
6
8
210
Die Primfakultät einer Zahl ist das Produkt der Primzahlen kleiner oder gleich der Zahl:
n
#
=
∏
p
=
2
p
∈
P
n
p
{\displaystyle n_{\#}=\prod _{\scriptstyle p\,=\,2 \atop \scriptstyle p\,\in \,\mathbb {P} }^{n}p\quad }
n
!n
n
!n
1
0
5
44
2
1
6
265
3
2
7
1854
4
9
8
14833
Die vor allem in der Kombinatorik auftretende Subfakultät
!
n
=
n
!
⋅
∑
k
=
0
n
(
−
1
)
k
k
!
{\displaystyle !n=n!\cdot \sum _{k=0}^{n}{\frac {(-1)^{k}}{k!}}}
bezeichnet die Anzahl aller fixpunktfreien Permutationen von
n
{\displaystyle n}
Elementen.
n
n
n
n
1
1
5
15
2
2
6
48
3
3
7
105
4
8
8
384
Die seltener verwendete Doppelfakultät oder doppelte Fakultät ist für gerade
n
{\displaystyle n}
das Produkt aller geraden Zahlen kleiner gleich
n
{\displaystyle n}
. Für ungerade
n
{\displaystyle n}
ist es das Produkt aller ungeraden Zahlen kleiner gleich
n
{\displaystyle n}
.
Sie ist definiert als:[9]
n
!
!
=
{
n
⋅
(
n
−
2
)
⋅
(
n
−
4
)
⋯
2
für
n
gerade und
n
>
0
,
n
⋅
(
n
−
2
)
⋅
(
n
−
4
)
⋯
1
für
n
ungerade und
n
>
0
,
1
für
n
∈
{
−
1
,
0
}
{\displaystyle n!!={\begin{cases}n\cdot (n-2)\cdot (n-4)\dotsm 2&{\text{für }}n{\text{ gerade und }}n>0,\\n\cdot (n-2)\cdot (n-4)\dotsm 1&{\text{für }}n{\text{ ungerade und }}n>0,\\1&{\text{für }}n\in \{-1,0\}\end{cases}}}
Häufig werden anstelle der Doppelfakultät Ausdrücke mit der gewöhnlichen Fakultät verwendet. Es gilt
(
2
k
)
!
!
=
2
k
k
!
{\displaystyle (2k)!!=2^{k}k!}
und
(
2
k
−
1
)
!
!
=
(
2
k
)
!
2
k
k
!
{\displaystyle (2k-1)!!={\frac {(2k)!}{2^{k}k!}}}
Werden nicht ganzzahlige Funktionswerte zugelassen, dann gibt es genau eine Erweiterung auf negative ungerade Zahlen, sodass
n
!
!
=
n
⋅
(
n
−
2
)
!
!
{\displaystyle n!!=n\cdot (n-2)!!}
für alle ungeraden ganzen Zahlen
n
{\displaystyle n}
gilt. Man erhält die Formel
n
!
!
=
1
n
+
2
⋅
1
n
+
4
⋯
1
1
{\displaystyle n!!={\tfrac {1}{n+2}}\cdot {\tfrac {1}{n+4}}\dotsm {\tfrac {1}{1}}}
für ungerade
n
<
0
{\displaystyle n<0}
.
Die Werte der Doppelfakultäten bilden die Folge A006882 in OEIS .
6
!
!
=
6
⋅
4
⋅
2
=
48
{\displaystyle 6!!=6\cdot 4\cdot 2=48}
7
!
!
=
7
⋅
5
⋅
3
⋅
1
=
105
{\displaystyle 7!!=7\cdot 5\cdot 3\cdot 1=105}
Die Anzahl
P
2
n
{\displaystyle P_{2n}}
der
n
{\displaystyle n}
-stelligen Kombinationen aus elementfremden Paaren gebildet aus
2
n
{\displaystyle 2n}
Elementen wird gegeben durch die Rekursion
P
2
n
=
(
2
n
−
1
)
P
2
n
−
2
{\displaystyle P_{2n}=(2n-1)P_{2n-2}}
mit Rekursionsanfang
P
2
=
1
{\displaystyle P_{2}=1}
(2 Elemente!). Auflösung der Rekursion ergibt
P
2
n
=
(
2
n
−
1
)
!
!
{\displaystyle P_{2n}=(2n-1)!!}
. Sollen z. B.
2
n
{\displaystyle 2n}
Mannschaften durch Verlosung paarweise aufeinandertreffen, dann ist die Wahrscheinlichkeit, dass dabei zwei bestimmte gegeneinander spielen, gegeben durch
P
2
n
−
2
P
2
n
=
1
2
n
−
1
{\displaystyle {\frac {P_{2n-2}}{P_{2n}}}={\frac {1}{2n-1}}}
.
Die Anzahl der Elemente der Hyperoktaedergruppe
B
n
{\displaystyle B_{n}}
ist
(
2
n
−
1
)
!
!
{\displaystyle (2n-1)!!}
.
Die Anzahl der fixpunktfreien involutorischen Permutationen von
2
n
{\displaystyle 2n}
Elementen ist
(
2
n
)
!
!
{\displaystyle (2n)!!}
.
Das
2
n
{\displaystyle 2n}
-te Moment der Standardnormalverteilung ist
(
2
n
−
1
)
!
!
{\displaystyle (2n-1)!!}
.
Auch in Integraltafeln und Formeln für spezielle Funktionen tritt die Doppelfakultät auf.
Für natürliche
n
{\displaystyle n}
gilt
(
2
n
−
1
)
!
!
=
2
n
π
Γ
(
n
+
1
2
)
{\displaystyle (2n-1)!!={\frac {2^{n}}{\sqrt {\pi }}}\Gamma \left(n+{\frac {1}{2}}\right)}
.
Analog zur doppelten Fakultät wird eine dreifache (
n
!
!
!
{\displaystyle n!!!}
), vierfache (
n
!
!
!
!
{\displaystyle n!!!!}
), …,
k
{\displaystyle k}
-fache Fakultät (
n
!
(
k
)
{\displaystyle n!^{(k)}}
) rekursiv definiert als
n
!
(
k
)
:=
{
1
falls
n
=
0
n
falls
0
<
n
≤
k
n
(
n
−
k
)
!
(
k
)
falls
n
>
k
{\displaystyle n!^{(k)}:={\begin{cases}1&{\text{falls }}n=0\\n&{\text{falls }}0<n\leq k\\n(n-k)!^{(k)}&{\text{falls }}n>k\end{cases}}}
[10]
Superfakultät und Hyperfakultät Bearbeiten
Natürlicher Logarithmus der Fakultät Bearbeiten
Gegeben ist die oben genannte Summenreihe für den Logarithmus naturalis aus der Fakultät:
γ
x
+
ln
[
Π
(
x
)
]
=
∑
n
=
1
∞
[
x
n
−
ln
(
1
+
x
n
)
]
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\sum _{n=1}^{\infty }{\biggl [}{\frac {x}{n}}-\ln {\biggl (}1+{\frac {x}{n}}{\biggr )}{\biggr ]}}
Diese Formel geht dann durch Darstellung der soeben gezeigten Formel mittels Stammfunktion der geometrischen Reihe und anschließenden Einsatzes der Definition der Riemannschen Zetafunktion hervor:
γ
x
+
ln
[
Π
(
x
)
]
=
∑
m
=
1
∞
[
ζ
(
2
m
)
2
m
x
2
m
−
ζ
(
2
m
+
1
)
2
m
+
1
x
2
m
+
1
]
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\sum _{m=1}^{\infty }{\biggl [}{\frac {\zeta (2m)}{2m}}\,x^{2m}-{\frac {\zeta (2m+1)}{2m+1}}\,x^{2m+1}{\biggr ]}}
Für die Debyeschen Funktionen gilt:
∫
0
∞
x
z
exp
(
x
)
−
1
d
x
=
Π
(
z
)
ζ
(
z
+
1
)
{\displaystyle \int _{0}^{\infty }{\frac {x^{z}}{\exp(x)-1}}\,\mathrm {d} x=\Pi (z)\,\zeta (z+1)}
Basierend auf dieser Identität kann folgende Integralidentität für den Logarithmus naturalis der Fakultätsfunktion hervorgebracht werden:
γ
x
+
ln
[
Π
(
x
)
]
=
∫
0
∞
exp
(
−
x
y
)
+
x
y
−
1
y
[
exp
(
y
)
−
1
]
d
y
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\int _{0}^{\infty }{\frac {\exp(-xy)+xy-1}{y{\bigl [}\exp(y)-1{\bigr ]}}}\,\mathrm {d} y}
Diese nun gezeigte Gleichung kommt auch durch Bildung der Ursprungsstammfunktion bezüglich x von folgender Formel hervor:
H
(
x
)
=
∫
0
∞
1
−
exp
(
−
x
y
)
exp
(
y
)
−
1
d
y
{\displaystyle \operatorname {H} (x)=\int _{0}^{\infty }{\frac {1-\exp(-xy)}{\exp(y)-1}}\,\mathrm {d} y}
Aus der gezeigten Formel kann das Element der Mascheroni-Konstante so entfernt werden:
ln
[
Π
(
x
)
]
=
∫
0
∞
1
y
{
x
exp
(
−
y
)
−
1
−
exp
(
−
x
y
)
exp
(
y
)
−
1
}
d
y
{\displaystyle \ln {\bigl [}\Pi (x){\bigr ]}=\int _{0}^{\infty }{\frac {1}{y}}{\biggl \{}x\exp(-y)-{\frac {1-\exp(-xy)}{\exp(y)-1}}{\biggr \}}\,\mathrm {d} y}
Für nähere Herleitungen siehe den Artikel Euler-Mascheroni-Konstante .
Der Begriff Superfakultät
sf
(
n
)
{\displaystyle \operatorname {sf} (n)}
wird für (wenigstens) zwei unterschiedliche Funktionen verwendet;[11]
Schwerpunktmäßig jedoch wird die Superfakultät als das Produkt der ersten
n
{\displaystyle n}
Fakultäten grundlegend definiert:
sf
(
n
)
=
∏
i
=
1
n
i
!
=
1
!
⋅
2
!
⋅
3
!
⋅
4
!
⋯
n
!
=
G
(
n
+
2
)
{\displaystyle \operatorname {sf} (n)=\prod _{i=1}^{n}i!=1!\cdot 2!\cdot 3!\cdot 4!\ \dotsm \ n!=G(n+2)}
[11] Die Funktion
G
(
z
)
{\displaystyle G(z)}
stellt die Barnessche G-Funktion dar.
Für alle natürlichen Zahlen identisch mit der soeben genannten Definition sind diese beiden Definitionen, welche die Superfakultät für alle
x
∈
R
∖
Z
<
0
{\displaystyle x\in \mathbb {R} \setminus \mathbb {Z} _{<0}}
definieren:
sf
(
x
)
=
Π
(
x
)
exp
{
x
2
[
ln
(
2
π
)
−
γ
x
−
x
−
1
]
}
∏
n
=
1
∞
[
(
1
+
x
n
)
n
exp
(
x
2
2
n
−
x
)
]
{\displaystyle \operatorname {sf} (x)=\Pi (x)\exp {\biggl \{}{\frac {x}{2}}{\bigl [}\ln(2\,\pi )-\gamma \,x-x-1{\bigr ]}{\biggr \}}\prod _{n=1}^{\infty }{\biggl [}{\bigl (}1+{\frac {x}{n}}{\bigr )}^{n}\exp {\bigl (}{\frac {x^{2}}{2n}}-x{\bigr )}{\biggr ]}}
sf
(
x
)
=
Γ
(
x
+
1
)
exp
{
x
2
[
ln
(
2
π
)
−
γ
x
−
x
−
1
]
}
∏
n
=
1
∞
[
(
1
+
x
n
)
n
exp
(
x
2
2
n
−
x
)
]
{\displaystyle \operatorname {sf} (x)=\Gamma (x+1)\exp {\biggl \{}{\frac {x}{2}}{\bigl [}\ln(2\,\pi )-\gamma \,x-x-1{\bigr ]}{\biggr \}}\prod _{n=1}^{\infty }{\biggl [}{\bigl (}1+{\frac {x}{n}}{\bigr )}^{n}\exp {\bigl (}{\frac {x^{2}}{2n}}-x{\bigr )}{\biggr ]}}
Für die Superfakultät sollen im Folgenden einige Werte genannt werden:[12]
sf
(
0
)
=
1
{\displaystyle \operatorname {sf} (0)=1}
sf
(
1
)
=
1
{\displaystyle \operatorname {sf} (1)=1}
sf
(
2
)
=
2
{\displaystyle \operatorname {sf} (2)=2}
sf
(
3
)
=
12
{\displaystyle \operatorname {sf} (3)=12}
sf
(
4
)
=
288
{\displaystyle \operatorname {sf} (4)=288}
sf
(
5
)
=
34560
{\displaystyle \operatorname {sf} (5)=34560}
Diese Werte sind in der Online-Enzyklopädie der Zahlenfolge unter dem Code OEIS: A000178 eingetragen.
Von anderen Mathematikern wurde der Begriff Superfakultät auch als mehrfache Potenz einer Fakultät verwendet:
n
$
=
n
!
n
!
⋅
⋅
⋅
n
!
⏟
n
!
{\displaystyle n\$=\underbrace {n!^{{n!}^{{\,\cdot }^{{\cdot }^{{\cdot \,}^{n!}}}}}} _{n!}}
Die Hyperfakultät
hf
(
n
)
{\displaystyle \operatorname {hf} (n)}
[13] ist für natürliche
n
{\displaystyle n}
folgendermaßen definiert:
hf
(
n
)
=
∏
i
=
1
n
i
i
=
1
1
⋅
2
2
⋅
3
3
⋅
4
4
⋯
n
n
{\displaystyle \operatorname {hf} (n)=\prod _{i=1}^{n}i^{i}=1^{1}\cdot 2^{2}\cdot 3^{3}\cdot 4^{4}\dotsm n^{n}}
[14] Sie kann durch die stochastische K-Funktion auf komplexe Zahlen verallgemeinert werden.
Es gilt hierfür folgende Formel:
hf
(
x
)
=
K
(
x
+
1
)
{\displaystyle \operatorname {hf} (x)=K(x+1)}
Hierbei sollte diese K-Funktion keinesfalls mit dem elliptischen K-Integral verwechselt werden.
So gelten diese Definitionsformeln für die Hyperfakultät in Abhängigkeit von der Gaußschen Pifunktion beziehungsweise Eulerschen Gammafunktion:
hf
(
x
)
=
exp
{
x
2
[
x
+
1
−
ln
(
2
π
)
]
+
∫
0
1
x
ln
[
Π
(
x
y
)
]
d
y
}
{\displaystyle \operatorname {hf} (x)=\exp {\biggl \{}{\frac {x}{2}}{\bigl [}x+1-\ln(2\,\pi ){\bigr ]}+\int \limits _{0}^{1}x\ln {\bigl [}\Pi (xy){\bigr ]}\,\mathrm {d} y{\biggr \}}}
hf
(
x
)
=
exp
{
x
2
[
x
+
1
−
ln
(
2
π
)
]
+
∫
0
1
x
ln
[
Γ
(
x
y
+
1
)
]
d
y
}
{\displaystyle \operatorname {hf} (x)=\exp {\biggl \{}{\frac {x}{2}}{\bigl [}x+1-\ln(2\,\pi ){\bigr ]}+\int \limits _{0}^{1}x\ln {\bigl [}\Gamma (xy+1){\bigr ]}\,\mathrm {d} y{\biggr \}}}
Basierend auf den genannten Definitionen gilt somit folgende Beziehung zwischen Hyperfakultät und Superfakultät:
hf
(
x
)
sf
(
x
)
=
(
x
!
)
x
+
1
=
Π
(
x
)
x
+
1
=
Γ
(
x
+
1
)
x
+
1
{\displaystyle \operatorname {hf} (x)\operatorname {sf} (x)=(x!)^{x+1}=\Pi (x)^{x+1}=\Gamma (x+1)^{x+1}}
Für die Hyperfakultät sollen im Folgenden einige Werte genannt werden:[15]
hf
(
0
)
=
1
{\displaystyle \operatorname {hf} (0)=1}
hf
(
1
)
=
1
{\displaystyle \operatorname {hf} (1)=1}
hf
(
2
)
=
4
{\displaystyle \operatorname {hf} (2)=4}
hf
(
3
)
=
108
{\displaystyle \operatorname {hf} (3)=108}
hf
(
4
)
=
27648
{\displaystyle \operatorname {hf} (4)=27648}
hf
(
5
)
=
86400000
{\displaystyle \operatorname {hf} (5)=86400000}
Diese Werte sind in der Online-Enzyklopädie der Zahlenfolge unter dem Code OEIS: A002109 eingetragen.
Dies sind die Gleichungen, die im Abschnitt Harmonische Reihenfunktion genannt wurden:
γ
x
+
ln
[
Π
(
x
)
]
=
∑
n
=
1
∞
[
x
n
−
ln
(
1
+
x
n
)
]
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\sum _{n=1}^{\infty }{\biggl [}{\frac {x}{n}}-\ln {\biggl (}1+{\frac {x}{n}}{\biggr )}{\biggr ]}}
γ
x
+
ln
[
Π
(
x
)
]
=
∑
m
=
1
∞
[
ζ
(
2
m
)
2
m
x
2
m
−
ζ
(
2
m
+
1
)
2
m
+
1
x
2
m
+
1
]
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\sum _{m=1}^{\infty }{\biggl [}{\frac {\zeta (2m)}{2m}}\,x^{2m}-{\frac {\zeta (2m+1)}{2m+1}}\,x^{2m+1}{\biggr ]}}
γ
x
+
ln
[
Π
(
x
)
]
=
∫
0
∞
exp
(
−
x
y
)
+
x
y
−
1
y
[
exp
(
y
)
−
1
]
d
y
{\displaystyle \gamma \,x+\ln {\bigl [}\Pi (x){\bigr ]}=\int _{0}^{\infty }{\frac {\exp(-xy)+xy-1}{y{\bigl [}\exp(y)-1{\bigr ]}}}\,\mathrm {d} y}
Durch Bildung der Ursprungsstammfunktion bezüglich x bei diesen Formeln entstehen nun folgenden Formeln:
x
2
[
γ
x
−
x
−
1
+
ln
(
2
π
)
]
+
ln
[
hf
(
x
)
]
=
∑
n
=
1
∞
[
x
2
2
n
−
(
n
+
x
)
ln
(
1
+
x
n
)
]
{\displaystyle {\frac {x}{2}}{\bigl [}\gamma \,x-x-1+\ln(2\,\pi ){\bigr ]}+\ln {\bigl [}\operatorname {hf} (x){\bigr ]}=\sum _{n=1}^{\infty }{\biggl [}{\frac {x^{2}}{2n}}-(n+x)\ln {\biggl (}1+{\frac {x}{n}}{\biggr )}{\biggr ]}}
x
2
[
γ
x
−
x
−
1
+
ln
(
2
π
)
]
+
ln
[
hf
(
x
)
]
=
∑
m
=
1
∞
[
ζ
(
2
m
)
2
m
(
2
m
+
1
)
x
2
m
+
1
−
ζ
(
2
m
+
1
)
(
2
m
+
1
)
(
2
m
+
2
)
x
2
m
+
2
]
{\displaystyle {\frac {x}{2}}{\bigl [}\gamma \,x-x-1+\ln(2\,\pi ){\bigr ]}+\ln {\bigl [}\operatorname {hf} (x){\bigr ]}=\sum _{m=1}^{\infty }{\biggl [}{\frac {\zeta (2m)}{2m(2m+1)}}\,x^{2m+1}-{\frac {\zeta (2m+1)}{(2m+1)(2m+2)}}\,x^{2m+2}{\biggr ]}}
x
2
[
γ
x
−
x
−
1
+
ln
(
2
π
)
]
+
ln
[
hf
(
x
)
]
=
∫
0
∞
x
2
y
2
−
2
x
y
+
2
−
2
exp
(
−
x
y
)
2
y
2
[
exp
(
y
)
−
1
]
d
y
{\displaystyle {\frac {x}{2}}{\bigl [}\gamma \,x-x-1+\ln(2\,\pi ){\bigr ]}+\ln {\bigl [}\operatorname {hf} (x){\bigr ]}=\int _{0}^{\infty }{\frac {x^{2}y^{2}-2\,xy+2-2\exp(-xy)}{2\,y^{2}{\bigl [}\exp(y)-1{\bigr ]}}}\,\mathrm {d} y}
Glaisher-Kinkelin-Konstante Bearbeiten
Die Superfakultät und die Hyperfakultät werden zur Definition der Glaisher-Kinkelin-Konstante angewendet:
A
=
lim
n
→
∞
n
!
sf
(
n
)
exp
[
(
1
2
n
2
−
1
12
)
ln
(
n
)
+
1
2
n
ln
(
2
π
)
−
3
4
n
2
+
1
12
]
{\displaystyle A=\lim _{n\to \infty }{\frac {n!}{\operatorname {sf} (n)}}\exp {\bigl [}{\bigl (}{\frac {1}{2}}n^{2}-{\frac {1}{12}}{\bigr )}\ln(n)+{\frac {1}{2}}n\ln(2\,\pi )-{\frac {3}{4}}n^{2}+{\frac {1}{12}}{\bigr ]}}
A
=
lim
n
→
∞
hf
(
n
)
exp
[
(
1
2
n
2
+
1
2
n
+
1
12
)
ln
(
n
)
−
1
4
n
2
]
{\displaystyle A=\lim _{n\to \infty }{\frac {\operatorname {hf} (n)}{\exp {\bigl [}{\bigl (}{\tfrac {1}{2}}n^{2}+{\tfrac {1}{2}}n+{\tfrac {1}{12}}{\bigr )}\ln(n)-{\tfrac {1}{4}}n^{2}{\bigr ]}}}}
Diese beiden genannten Definitionen stimmen miteinander überein.
Ebenso mit diesen Definitionen übereinstimmend ist diese Abel-Plana-Integraldarstellung für die Glaisher-Kinkelin-Konstante:
A
=
exp
[
1
3
−
∫
0
∞
2
arctan
(
x
)
+
x
ln
(
x
2
+
1
)
2
exp
(
π
x
)
sinh
(
π
x
)
d
x
]
{\displaystyle A=\exp {\biggl [}{\frac {1}{3}}-\int _{0}^{\infty }{\frac {2\arctan(x)+x\ln(x^{2}+1)}{2\exp(\pi \,x)\sinh(\pi \,x)}}\,\mathrm {d} x{\biggr ]}}
A
≈
1,282
42712
91006226
36875342
56886979
17277676
88927325
{\displaystyle A\approx 1{,}28242712\,\,\,91006226\,\,\,36875342\,\,\,56886979\,\,\,17277676\,\,\,88927325}
Und die Glaisher-Kinkelin-Konstante findet beispielsweise bei den Riemannschen und Dirichletschen Funktionen Anwendung:
Abszissenwerte x
Riemannsche Zetaableitung
Dirichletsche Lambdaableitung
Dirichletsche Etaableitung
+
2
{\displaystyle +2}
ζ
′
(
2
)
=
−
1
6
π
2
[
12
ln
(
A
)
−
ln
(
2
π
)
−
γ
]
{\displaystyle \zeta '(2)=-{\tfrac {1}{6}}\pi ^{2}{\bigl [}12\ln(A)-\ln(2\pi )-\gamma {\bigr ]}}
λ
′
(
2
)
=
−
1
24
π
2
[
36
ln
(
A
)
−
ln
(
16
π
3
)
−
3
γ
]
{\displaystyle \lambda '(2)=-{\tfrac {1}{24}}\pi ^{2}{\bigl [}36\ln(A)-\ln(16\pi ^{3})-3\,\gamma {\bigr ]}}
η
′
(
2
)
=
1
12
π
2
[
ln
(
4
π
)
−
12
ln
(
A
)
+
γ
]
{\displaystyle \eta '(2)={\tfrac {1}{12}}\pi ^{2}{\bigl [}\ln(4\pi )-12\ln(A)+\gamma {\bigr ]}}
−
1
{\displaystyle -1}
ζ
′
(
−
1
)
=
1
12
−
ln
(
A
)
{\displaystyle \zeta '(-1)={\tfrac {1}{12}}-\ln(A)}
λ
′
(
−
1
)
=
ln
(
A
)
−
1
6
ln
(
2
)
−
1
12
{\displaystyle \lambda '(-1)=\ln(A)-{\tfrac {1}{6}}\ln(2)-{\tfrac {1}{12}}}
η
′
(
−
1
)
=
3
ln
(
A
)
−
1
3
ln
(
2
)
−
1
4
{\displaystyle \eta '(-1)=3\ln(A)-{\tfrac {1}{3}}\ln(2)-{\tfrac {1}{4}}}
Leonhard Euler: Remarques sur un beau rapport entre les séries des puissances tant directes que réciproques. (1749), Histoire de l’Académie Royale des Sciences et Belles-Lettres 17 (1761), 1768, S. 96/97 (französisch).
Leonhard Euler: Evolutio formulae integralis ∫ x f − 1 d x ( l x ) m n {\displaystyle \textstyle \int \!x^{f-1}\mathrm {d} x(lx)^{\frac {m}{n}}} integratione a valore x=0 ad x=1 extensa. 4. Juli 1771, Novi commentarii academiae scientiarum imperialis Petropolitanae 16, 1772, S. 121 (lateinisch).
Adrien-Marie Legendre: Recherches sur diverses sortes d’intégrales définies. (13. November 1809), Mémoires de la classe des sciences mathématiques et physiques de l’Institut de France 10, 1809, S. 485 (französisch).
Hermann Kinkelin : Ueber eine mit der Gammafunction verwandte Transcendente und deren Anwendung auf die Integralrechnung . (Juli 1856) In: Journal für die reine und angewandte Mathematik , 57, 1860, S. 122–138 (beim GDZ: digizeitschriften.de )
J. W. L. Glaisher : On the Product 1¹.2².3³...nⁿ . In: The Messenger of Mathematics , 7, 1878, S. 43–47 (englisch; „A =1·28242 7130“ auf S. 43); Textarchiv – Internet Archive
↑ V. H. Moll: Some Questions in the Evaluation of Definite Integrals. MAA Short Course, San Antonio, TX. Jan. 2006. Archivierte Kopie (Memento vom 2. April 2008 im Internet Archive )
↑ Königliche Akademie der Wissenschaften (Berlin): Histoire de l'Académie Royale des Sciences et des Belles Lettres de Berlin: depuis ... : avec les mémoires tirez des registres de cette Academie. 1761 (1768) . Haude & Spener, 1768 (google.de [abgerufen am 26. Januar 2023]).
↑ Imperatorskai︠a︡ akademīi︠a︡ naukʺ i khudozhestvʺ (Russia): Novi commentarii Academiae Scientiarum Imperialis Petropolitanae . Typis Academiae Scientarum, 1772 (google.de [abgerufen am 26. Januar 2023]).
↑ Institut de France: Mémoires de l'Institut des Sciences et Arts . Paris, 1810 (archive.org [abgerufen am 26. Januar 2023]).
↑ Math Wonders To inspire Teachers and Students. (PDF) Abgerufen am 20. Januar 2023 .
↑ Stirling's Approximation. In: Wolfram MathWorld. Abgerufen am 20. Januar 2023 (englisch).
↑ Leonhard Euler : De progressionibus transcendentibus, seu quarum termini generales algebraice dari nequeunt. (28. November 1729), Commentarii academiae scientiarum imperialis Petropolitanae 5, 1738, S. 36–57 (lateinisch).
↑ E. Freitag, R. Busam: Funktionentheorie . Springer-Verlag, ISBN 3-540-31764-3 , S. 225 .
↑ Eric W. Weisstein : Double Factorial . In: MathWorld (englisch).
↑ Eric W. Weisstein : Multifactorial . In: MathWorld (englisch).
↑ a b Eric W. Weisstein : Superfactorial . In: MathWorld (englisch).
↑ A000178 – OEIS. Abgerufen am 9. Januar 2023 .
↑ Special factorials. 12. Oktober 2022, abgerufen am 8. Januar 2023 (englisch).
↑ Eric W. Weisstein : Hyperfactorial . In: MathWorld (englisch).
↑ A002109 - OEIS. Abgerufen am 9. Januar 2023 .