Arab Academy for Science, Technology & Maritime

Transport

Math 5 (BA323)
(Gamma Function)

Fall 2020-2021
Chapter 2 Gamma Function

Chapter 2

Gamma Function

Introduction

Gamma function is used to generalize the factorial function to nonintegral values, introduced by the
Swiss mathematician Leonhard Euler in the 18th century. The gamma function belongs to the category of
the special functions. It also appears in various area as asymptotic series, definite integration and
hypergeometric series.

Definitions of the gamma function

To extend the factorial to any real number 𝑥 > 0 (whether or not 𝑥 is a whole number), the gamma
function is defined as

∞
Γ(𝑥) = ∫ 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡.
0

By using techniques of integration, it can be shown that

∞
∞
Γ(1) = ∫ 𝑒 −𝑡 𝑑𝑡 = −𝑒 −𝑡 | 0 = 0 + 1 = 1 = 0!
0

Similarly, using a technique from calculus known as integration by parts, it can be proved that the gamma
function has the following recursive property: if 𝑥 > 0, then

∞
∞
Γ(2) = ∫ 𝑡 𝑒 −𝑡 𝑑𝑡 = (−𝑡𝑒 −𝑡 − 𝑒 −𝑡 ) | 0 = 0 + 1 = 1 = 1!,
0

∞
∞
Γ(3) = ∫ 𝑡 2 𝑒 −𝑡 𝑑𝑡 = (−𝑡 2 𝑒 −𝑡 − 2𝑡𝑒 −𝑡 − 2𝑒 −𝑡 ) | 0 = 0 + 2 = 2 = 2!,
0

-1-
Chapter 2 Gamma Function

and
∞
∞
Γ(4) = ∫ 𝑡 3 𝑒 −𝑡 𝑑𝑡 = (−𝑡 3 𝑒 −𝑡 − 3𝑡 2 𝑒 −𝑡 − 6𝑡𝑒 −𝑡 − 6𝑒 −𝑡 ) | 0 = 0 + 6 = 6 = 3!,
0

and so on. Generally, if 𝑥 is a natural number (1, 2, 3, … ), then

Γ(𝑥) = (𝑥 − 1)!

While the gamma function behaves like a factorial for natural numbers (a discrete set), its extension to the
positive real numbers (a continuous set) makes it useful for modeling situations involving continuous change,
with important applications to calculus, differential equations, complex analysis, and statistics.

Now, we want to prove that

Γ(1/2) = √𝜋.

To prove that, we remember the Gamma function

∞
Γ(𝑥) = ∫ 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡.
0
2
Making the substitution 𝑡 = 𝑢 gives the equivalent expression
∞
2
Γ(𝑥) = 2 ∫ 𝑢2𝑥−1 𝑒 −𝑢 𝑑𝑢.
0
1 1
A special value of the gamma function can be derived when 2𝑥 − 1 = 0 (𝑥 = 2). When (𝑥 = 2) 2 , Γ(1/2)
simplifies as
∞
2
Γ(1/2) = 2 ∫ 𝑒 −𝑢 𝑑𝑢.
0
To derive the value for Γ(1/2), the following steps are used. First, the value of Γ(1/2) is squared. Second,
the squared value is rewritten as a double integral. Third, the double integral is evaluated by transforming to
polar coordinates. Fourth, the Γ(1/2) is explicitly solved for. First, square the value for Γ(1/2) and rewrite
as a double integral.
1 2 ∞
2
∞
2
∞ ∞
2 2
[Γ ( )] = Γ(1/2)Γ(1/2) = [2 ∫ 𝑒 −𝑢 𝑑𝑢 ] [2 ∫ 𝑒 −𝑣 𝑑𝑣] = 4 ∫ ∫ 𝑒 −𝑢 −𝑣 𝑑𝑣 𝑑𝑢.
2 0 0 0 0

The region 𝑅 which defines the first quadrant, is the region of integration. The polar transformation 𝑢 𝑟 cos 𝜃,
𝑣 = 𝑟 sin 𝜃 will transform the integral problem from cartesian coordinates to polar coordinates, (𝑟, 𝜃). These
𝜋
new variables will range from 0 ≤ 𝑟 ≤ ∞ and 0 ≤ 𝜃 ≤ 2 for the first quadrant.

Hence,
𝜋 𝜋
1 2 ∞ ∞
2 2 2
∞
2 2 1 ∞
[Γ ( )] = 4 ∫ ∫ 𝑒 −𝑢 −𝑣 𝑑𝑣 𝑑𝑢 = 4 ∫ ∫ 𝑒 −𝑟 𝑟 𝑑𝑟 𝑑𝜃 = 4 ∫ (− 𝑒 −𝑡 ) | 0 𝑑𝜃
2 0 0 0 0 0 2
𝜋 𝜋
2
= 2 ∫ 𝑑𝜃 = 2𝜃 | 2 = 𝜋.
0 0
Hence,

1 2 1
[Γ ( )] = 𝜋 ⇒ Γ ( ) = √𝜋
2 2
-2-
Chapter 2 Gamma Function

Also, we can prove that

Γ(𝑥 + 1) = 𝑥 Γ(𝑥).
The Gamma function is
∞
Γ(𝑥) = ∫ 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡,
0
then,
∞
Γ(𝑥 + 1) = ∫ 𝑡 𝑥 𝑒 −𝑡 𝑑𝑡.
0
by applying the integration by parts yields,
∞ ∞ ∞
∞
Γ(𝑥 + 1) = ∫ 𝑡 𝑥 𝑒 −𝑡 𝑑𝑡 = (−𝑡 𝑥 𝑒 −𝑡 ) | 0 + 𝑥 ∫ 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡 = 𝑥 ∫ 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡 = 𝑥 Γ(𝑥)
0 0 0

Graph and Table of the Gamma function 𝚪(𝒙) in a real domain.

The following graph represents the Gamma function in real value of 𝑥. Remain that the value of Gamma
function at zero an negative integer equal ±∞, or

Γ(0) = Γ(−1) = Γ(−2) = ⋯ = ±∞

Because Γ(𝑥 + 1) = 𝑥 Γ(𝑥), we can write

Γ(𝑥 + 1)
Γ(𝑥) =
𝑥

So,
Γ(1) Γ(0) Γ(−1)
Γ(0) = = ±∞, Γ(−1) = = ±∞, Γ(−2) = = ±∞, ⋯
0 −1 −2

-3-
Chapter 2 Gamma Function

The following table shows Gamma function on the interval [1, 2].

Gamma Table 𝚪(𝒙), 𝟏 ≤ 𝒙 ≤ 𝟐

𝚪(𝒙) 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
1.0 1.00000 0.99432 0.98884 0.9835 0.97843 0.9735 0.96874 0.96415 0.95972 0.95545
1.1 0.95135 0.94739 0.9435 0.93993 0.93641 0.93304 0.9298 0.92669 0.92372 0.92088
1.2 0.91816 0.91557 0.91310 0.91075 0.90852 0.90640 0.90439 0.90250 0.90071 0.89904
1.3 0.89747 0.89600 0.89464 0.89337 0.89221 0.89115 0.89018 0.88931 0.88853 0.88785
1.4 0.88726 0.88676 0.88635 0.88603 0.8858 0.88566 0.88560 0.88563 0.88574 0.88594
1.5 0.88622 0.88659 0.88703 0.88756 0.88817 0.88886 0.88963 0.89049 0.89142 0.89242
1.6 0.89351 0.89468 0.89592 0.89724 0.89864 0.90011 0.90166 0.90329 0.90500 0.90678
1.7 0.90863 0.91057 0.91258 0.91466 0.91682 0.91906 0.92137 0.92376 0.92622 0.92876
1.8 0.93138 0.93407 0.93684 0.93969 0.94261 0.94561 0.94868 0.95184 0.95507 0.95837
1.9 0.96176 0.96523 0.96877 0.97239 0.97609 0.97988 0.98374 0.98768 0.99170 0.99581

Example 1 Find the values of the following Gamma functions

𝟑 𝟑 𝟓 𝟓
𝚪 (𝟐) , 𝚪 (− 𝟐) , 𝚪 (𝟐), 𝚪 (− 𝟐) , 𝚪(𝟑. 𝟑𝟓), 𝚪(−𝟐. 𝟓𝟐), 𝚪(𝟓. 𝟕𝟗), 𝚪(−𝟏. 𝟓𝟒),
Solution:

3 1 1 1
Γ ( ) = Γ ( ) = √𝜋.
2 2 2 2

1 1
3 Γ (− 2) Γ (2) √𝜋 4 √ 𝜋
Γ (− ) = = = = .
2 3 3 1 3 3
−2 −2 × −2 4

5 3 3 3 1 1 3 1 3
Γ ( ) = Γ ( ) = × Γ ( ) = × √𝜋 = √𝜋.
2 2 2 2 2 2 2 2 4

3 1 1
5 Γ (− 2) Γ (− 2) Γ (2) √𝜋 8√𝜋
Γ (− ) = = = = =− .
2 5 5 3 5 3 1 15 15
−2 −2 × −2 −2 × −2 × −2 − 8

Γ(3.35) = 2.35 Γ(2.35) = 2.35 × 1.35 Γ(1.35) = 2.35 × 1.35 × 0.89115 (from Gamma Table)
= 2.35 Γ(2.35) = 2.8271.

Γ(−1.52) Γ(−0.52) Γ(0.48)

Γ(−2.52) = = =
−2.52 −2.52 × −1.52 −2.52 × −1.52 × −0.52
Γ(1.48) 0.88574
= = (from Gamma Table)
−2.52 × −1.52 × −0.52 × 0.48 −0.95606

= 2.35 Γ(2.35) = 2.8271.

-4-
Chapter 2 Gamma Function

Γ(5.79) = 4.79 Γ(4.79) = 4.79 × 3.79 Γ(3.79) = 4.79 × 3.79 × 2.79 Γ(2.79)
= 4.79 × 3.79 × 2.79 Γ(2.79) = 4.79 × 3.79 × 2.79 × 1.79 Γ(1.79)
= 4.79 × 3.79 × 2.79 × 1.79 × 0.92876 (from Gamma Table)
= 84.2045.

Γ(−0.54) Γ(0.46) Γ(1.46)

Γ(−1.54) = = ==
−1.54 −1.54 × −0.54 −1.54 × −0.54 × 0.46
0.88574
= (from Gamma Table)
0.382536

= 2.31508.

Example 2 Find the following integrals using Gamma Function

∞ ∞ ∞ 𝟑
(𝐢) ∫𝟎 𝒙𝟓 𝒆−𝒙 𝒅𝒙 (𝐢𝐢) ∫𝟎 𝒙𝟑 𝒆𝟏𝟎−𝟐𝒙 𝒅𝒙 (𝐢𝐢𝐢) ∫𝟎 𝟏𝟎−𝟐𝒙 𝒅𝒙
Solution:
(i)
∞
∫ 𝑥 5 𝑒 −𝑥 𝑑𝑥 = Γ(6) = 5! = 120.
0

(ii)
∞ ∞
∫ 𝑥 3 𝑒 10−2𝑥 𝑑𝑥 = 𝑒 10 ∫ 𝑥 3 𝑒 −2𝑥 𝑑𝑥
0 0

To convert the resulted integral to Gamma function, we use the substitution

𝑡 1
𝑡 = 2𝑥 ⇒ 𝑥 = ⇒ 𝑑𝑥 = 𝑑𝑡
2 2
At 𝑥 = 0 ⇒ 𝑡 = 0, at 𝑥 = ∞ ⇒ 𝑡 = ∞
So, we can write the previous integral as

∞ ∞
𝑒 10 ∞ 𝑡 3 −𝑡 𝑒 10 ∞ 3 −𝑡 𝑒 10
∫ 𝑥 3 𝑒 10−2𝑥 𝑑𝑥 = 𝑒 10 ∫ 𝑥 3 𝑒 −2𝑥 𝑑𝑥 = ∫ ( ) 𝑒 𝑑𝑡 = ∫ 𝑡 𝑒 𝑑𝑡 = Γ(4)
0 0 2 0 2 16 0 16

𝑒 10 3𝑒 10
= × 3! = .
16 8
(iii)
∞ ∞ 3
∞
−2𝑥 3 ln 10−2𝑥 3 ln 10
∫ 10 𝑑𝑥 = ∫ 𝑒 𝑑𝑥 = ∫ 𝑒 −2𝑥 𝑑𝑥
0 0 0

To convert the resulted integral to Gamma function, we use the substitution

2
−
3
𝑡 3 1 𝑡 3
𝑡 = 2𝑥 ln 10 ⇒ 𝑥 = √ ⇒ 𝑑𝑥 = ( ) 𝑑𝑡
2 ln 10 6 ln 10 2 ln 10
At 𝑥 = 0 ⇒ 𝑡 = 0, at 𝑥 = ∞ ⇒ 𝑡 = ∞
So, we can write the previous integral as

2
∞ ∞ ∞ ∞ −
−2𝑥 3 ln 10−2𝑥
3
−2𝑥 3 ln 10
1 𝑡 3
∫ 10 𝑑𝑥 = ∫ 𝑒 𝑑𝑥 = ∫ 𝑒 𝑑𝑥 = ∫ 𝑒 −t ( ) 𝑑𝑡
0 0 0 0 6 ln 10 2 ln 10
-5-
Chapter 2 Gamma Function
2 2
∞ − ∞ −
−2𝑥 3
1 1 3
−
2
−t
1 1 3 1
∫ 10 𝑑𝑥 = ( ) ∫ 𝑡 𝑒 𝑑𝑡 =
3 ( ) Γ( )
0 6 ln 10 2 ln 10 0 6 ln 10 2 ln 10 3

−
2 4 −
2
1 1 3 Γ (3) 1 1 3 0.89337
= ( ) × = ( ) × = 0.53696 .
6 ln 10 2 ln 10 1 6 ln 10 2 ln 10 1
3 3

Example 2 Find the following integrals using Gamma Function

𝟏 𝟏 𝟏 𝟑 𝟏 𝟏 𝟐
(𝐢) ∫𝟎 √− 𝐥𝐧 𝒙 𝒅𝒙 (𝐢𝐢) ∫𝟎 (𝐥𝐧 ) 𝒅𝒙 (𝐢𝐢𝐢) ∫𝟎 √𝒙𝟐 − 𝟐𝒙 + 𝟏 (𝐥𝐧 ) 𝒅𝒙
𝒙 √𝟏−𝒙
Solution:
(i)
1
∫ √− ln 𝑥 𝑑𝑥
0

To convert the resulted integral to Gamma function, we use the substitution

2 2
𝑡 = √− ln 𝑥 ⇒ 𝑥 = 𝑒 −𝑡 ⇒ 𝑑𝑥 = −2𝑡 𝑒 −𝑡 𝑑𝑡
At 𝑥 = 0 ⇒ 𝑡 = ∞, at 𝑥 = 1 ⇒ 𝑡 = 0
So, we can write the previous integral as

1 0 ∞
2 2
∫ √− ln 𝑥 𝑑𝑥 = ∫ 𝑡 (−2𝑡 𝑒 −𝑡 )𝑑𝑡 = 2 ∫ 𝑡 2 𝑒 −𝑡 𝑑𝑡
0 ∞ 0

We need another substitution to get the pervious integral in Gamma function

1 1
𝑢 = 𝑡 2 ⇒ 𝑡 = √𝑢 ⇒ 𝑑𝑡 = 𝑑𝑢
2 √𝑢
At 𝑡 = 0 ⇒ 𝑢 = 0, at 𝑡 = ∞ ⇒ 𝑢 = ∞
1 ∞ ∞ ∞ 1
2 2 1 3 1
∫ √− ln 𝑥 𝑑𝑥 = 2 ∫ 𝑡 2 𝑒 −𝑡 𝑑𝑡 = ∫ (√𝑢) 𝑒 −𝑢 𝑑𝑢 = ∫ 𝑢2 𝑒 −𝑢 𝑑𝑢 = Γ ( ) = √𝜋
0 0 0 √𝑢 0 2 2

(ii)
1 3
1
∫ (ln ) 𝑑𝑥
0 𝑥
To convert the resulted integral to Gamma function, we use the substitution
1
𝑡 = ln ⇒ 𝑥 = 𝑒 −𝑡 ⇒ 𝑑𝑥 = − 𝑒 −𝑡 𝑑𝑡
𝑥
At 𝑥 = 0 ⇒ 𝑡 = ∞, at 𝑥 = 1 ⇒ 𝑡 = 0
So, we can write the previous integral as

1 3 0 ∞
1
∫ (ln ) 𝑑𝑥 = − ∫ 𝑡 𝑒 𝑑𝑡 = ∫ 𝑡 3 𝑒 −𝑡 𝑑𝑡 = Γ(4) = 3! = 6
3 −𝑡
0 𝑥 ∞ 0

-6-
Chapter 2 Gamma Function
(iii)
1 2
1
∫ √𝑥 2 − 2𝑥 + 1 (ln ) 𝑑𝑥
0 √1 − 𝑥
To convert the resulted integral to Gamma function, we use the substitution
1
𝑡 = ln ⇒ 𝑥 = 1 − 𝑒 −2𝑡 ⇒ 𝑑𝑥 = −2 𝑒 −2𝑡 𝑑𝑡
√1 − 𝑥
At 𝑥 = 0 ⇒ 𝑡 = 0, at 𝑥 = 1 ⇒ 𝑡 = ∞

So, we can write the previous integral as

1 2 1 2
1 1
∫ √𝑥 2 − 2𝑥 + 1 (ln ) 𝑑𝑥 = ∫ √(𝑥 − 1)2 (ln ) 𝑑𝑥
0 √1 − 𝑥 0 √1 − 𝑥
∞ ∞
−2𝑡 2 (−2 −2𝑡 )𝑑𝑡
= ∫ −𝑒 𝑡 𝑒 = 2 ∫ 𝑡 2 𝑒 −4𝑡 𝑑𝑡
0 0

We need another substitution to get the pervious integral in Gamma function form
𝑢 1
𝑢 = 4𝑡 ⇒ 𝑡 = ⇒ 𝑑𝑡 = 𝑑𝑢
4 4
At 𝑡 = 0 ⇒ 𝑢 = 0, at 𝑡 = ∞ ⇒ 𝑢 = ∞
1 2 ∞ ∞
1 𝑢 2 1 1 ∞ 2 −𝑢
∫ √𝑥 2 − 2𝑥 + 1 (ln ) 𝑑𝑥 = 2 ∫ 𝑡 2 𝑒 −4𝑡 𝑑𝑡 = 2 ∫ ( ) 𝑒 −𝑢 ( ) 𝑑𝑢 = ∫ 𝑢 𝑒 𝑑𝑢
0 √1 − 𝑥 0 0 4 4 32 0
1 1
= Γ(3) =
32 16

Example 3 Prove that

𝟏 (𝟐𝒏−𝟏)!
𝚪 (𝒏 + 𝟐) = 𝟐𝒏−𝟏 √𝝅
𝟐 (𝒏−𝟏)!
Solution:
1 1 1 1 3 3
Γ (𝑛 + ) = (𝑛 − ) Γ (𝑛 − ) = (𝑛 − ) (𝑛 − ) Γ (𝑛 − )
2 2 2 2 2 2
1 3 5 5
= (𝑛 − ) (𝑛 − ) (𝑛 − ) Γ (𝑛 − )
2 2 2 2
1 3 5 3 1 1
= (𝑛 − ) (𝑛 − ) (𝑛 − ) ⋯ ( ) ( ) Γ ( )
2 2 2 2 2 2
2𝑛 − 1 2𝑛 − 3 2𝑛 − 5 3 1
=( )( )( ) ⋯ ( ) ( ) √𝜋
2 2 2 2 2
(2𝑛 − 1)(2𝑛 − 3)(2𝑛 − 5) ⋯ (3)(1)
= √𝜋
2𝑛
(2𝑛 − 1)(2𝑛 − 2)(2𝑛 − 3)(2𝑛 − 4)(2𝑛 − 5) ⋯ (3)(1)
= √𝜋
2𝑛 (2𝑛 − 2)(2𝑛 − 4) ⋯ (2)
(2𝑛 − 1)(2𝑛 − 2)(2𝑛 − 3)(2𝑛 − 4)(2𝑛 − 5) ⋯ (3)(1)
= √𝜋
2𝑛 2𝑛−1 (𝑛 − 1)(𝑛 − 2) ⋯ (1)
(2𝑛 − 1)!
= √𝜋.
22𝑛−1 (𝑛 − 1)!
-7-
Chapter 2 Gamma Function
Sheet 2
Beta and Gamma Functions

1. Evaluate
7 7 4 Γ(7)
(i) Γ ( ) , Γ (− ) , Γ ( ) , Γ(2.7) (ii)
2 2 3 Γ(4)Γ(6)
3
1 3 5 Γ(3) Γ ( )
(iii) Γ ( ) Γ ( ) Γ ( ) (iv) 2
2 2 2 9
Γ( )
2
2. Find the following integrals using Gamma Function
∞ ∞ 1
(i) ∫ 𝑥 3 𝑒 −𝑥 𝑑𝑥 (ii) ∫ 𝑥 2 𝑒 2−5𝑥 𝑑𝑥 (iii) ∫ √− ln 𝑥 𝑑𝑥
0 0 0
2
∞
−𝑥 4
∞
−3𝑥 2
1 𝑥
1
(iv) ∫ 𝑒 𝑑𝑥 (v) ∫ 4 𝑑𝑥 (vi) ∫ ln ( ) 𝑑𝑥
0 0 0 𝑥
𝜋
∞ 1
2
(vii) ∫ 𝑒 (𝑥−𝑒 𝑥 ) (viii) ∫ (tan 𝑥 + tan 𝑥 )𝑒 3 5 − tan2 𝑥 (ix) ∫ √ln(1/𝑥) 𝑑𝑥
𝑑𝑥 𝑑𝑥
−∞ 0 0
∞ 1−𝑥 2 1 ∞
𝑥𝑒 1 1
(x) ∫ 𝑑𝑥 (xi) ∫ (𝑥 − 1)2 ln 𝑑𝑥 (xii) ∫ 3 𝑑𝑥
1 √𝑥 2 − 1 0 (1 − 𝑥 ) 2 1 𝑥 2 √ln 𝑥
3. Find the following integrals Beta function
1 1 4
3( 4
(i) ∫ 𝑥 1 − 𝑥 )2 𝑑𝑥 (ii) ∫ √(1/𝑥) − 1 𝑑𝑥 (iii) ∫ 𝑥 2 √4 − 𝑥 𝑑𝑥
0 0 0
3 𝜋
∞ 3
1 𝑥2 2
(iv) ∫ 𝑑𝑥 (v) ∫ 3 𝑑𝑥 (vi) ∫ cos4 𝜃 𝑑𝜃
0 √𝑥(1 + 𝑥) 0 √27 − 𝑥3 0
𝜋 𝜋
5 5 2
2 2
(vii) ∫ tan3/2 𝜃 𝑑𝜃 (viii) ∫ sin2 𝜃 cos2 𝜃 𝑑𝜃 (ix) ∫ 𝑥 3 √5 − 𝑥 𝑑𝑥
0 0 0
𝜋 𝜋
3 2
sin 𝜃
2 2
2
1
(x) ∫ 2
𝑑𝜃 (xi) ∫ cos (2𝜃) 𝑑𝜃 (xii) ∫ 𝑑𝑥
0 1 + tan 𝜃 0 0 √4 − 𝑥 2
𝜋
1 2
𝑥3 2
5 2
1
(xiii) ∫ 𝑑𝜃 (xiv) ∫ cos 𝜃 (sec 𝜃 − 1) 𝑑𝜃 (xv) ∫ 𝑥 (8 − 𝑥 3 )3 𝑑𝑥
0 √1 − 𝑥 2 0 0

4. Prove that the following

2
1 (Γ(𝑛))
(i) Β(𝑛, 𝑛 + 1) = .
2 Γ(2𝑛)
(ii) Β(𝑝, 𝑞)Β(𝑝 + 𝑞, 𝑚) = Β(𝑞, 𝑚)Β(𝑞 + 𝑚, 𝑝).
(iii) Β(𝑚, 𝑛) = Β(𝑚 + 1, 𝑛) + Β(𝑚, 𝑛 + 1) and then deduce that
𝑛
Β(𝑚, 𝑛 + 1) = Β(𝑚, 𝑛).
𝑚+𝑛

1
(iii) Β (𝑛, 𝑛 + ) = 22𝑛 Β(2𝑛, 2𝑛).
2
-8-