0% found this document useful (0 votes)
93 views9 pages

Beta and Gamma Functions

1) The Beta function is defined as the integral from 0 to 1 of x^m-1 (1-x)^n-1 dx, where m and n are positive. It is denoted as B(m,n). 2) The Gamma function is defined as the integral from 0 to infinity of e^-x x^n-1 dx, where n is positive. It is denoted as Γ(n). 3) These functions have various properties including symmetry, relationships between each other, and formulas for calculating their values.

Uploaded by

SaptarshiDutta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
93 views9 pages

Beta and Gamma Functions

1) The Beta function is defined as the integral from 0 to 1 of x^m-1 (1-x)^n-1 dx, where m and n are positive. It is denoted as B(m,n). 2) The Gamma function is defined as the integral from 0 to infinity of e^-x x^n-1 dx, where n is positive. It is denoted as Γ(n). 3) These functions have various properties including symmetry, relationships between each other, and formulas for calculating their values.

Uploaded by

SaptarshiDutta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Beta and Gamma Functions

If (i) the interval [a, b] is finite

(ii) the function f(x) is bounded in [a, b], that is, f(x) does not become infinite at any point in the
interval, and
𝑑 𝑏
(iii) 𝑑𝑥 [𝜙(𝑥)] = 𝑓(𝑥), then ∫𝑎 𝑓(𝑥)𝑑𝑥 = (b) - (a) is called a proper integral.

𝑏
In condition (i) is not satisfied (that is, a, b or both are infinite), the integral ∫𝑎 𝑓(𝑥)𝑑𝑥 is called
an improper integral of first kind.
1
Definition: If m and n are positive, then ∫0 𝑥 𝑚−1 (1 − 𝑥)𝑛−1 𝑑𝑥 is called a Beta function and
denoted by B(m, n) or (m, n).

Note: If m and n are both greater than or equal to 1, then the above integral is a proper integral.
On the other hand if either m or n is less than 1, then the integral becomes improper but may be
convergent.

Properties of Beta Functions:

1. Symmetry: B(m, n) = B(n, m)


1
Verification: By definition, B(m, n) = ∫0 𝑥 𝑚−1 (1 − 𝑥)𝑛−1 𝑑𝑥

1 𝑎 𝑎
= ∫0 (1 − 𝑥)𝑚−1 [(1 − (1 − 𝑥))]𝑛−1 𝑑𝑥 (since ∫0 𝑓(𝑥)𝑑𝑥 = ∫0 𝑓(𝑎 − 𝑥)𝑑𝑥)

1
= ∫0 (1 − 𝑥)𝑚−1 𝑥 𝑛−1 𝑑𝑥

1
= ∫0 𝑥 𝑛−1 (1 − 𝑥)𝑚−1 𝑑𝑥 = B(n, m).

𝜋
1 𝑝+1 𝑞+1
2. If p > -1, q > -1 then ∫02 𝑠𝑖𝑛𝑝 𝜃 𝑐𝑜𝑠 𝑞 𝜃 𝑑𝜃 = 2B( , ).
2 2

𝜋 𝜋 (𝑝−1)⁄ (𝑞−1)⁄
Verification: ∫02 𝑠𝑖𝑛𝑝 𝜃 𝑐𝑜𝑠 𝑞 𝜃 𝑑𝜃 = ∫02 (𝑠𝑖𝑛2 𝜃) 2 (𝑐𝑜𝑠 2 𝜃) 2 sin 𝜃 cos 𝜃 𝑑𝜃

Put sin2 = x. Then cos2 d = 1- x.


𝜋
When  = 0, we have x = 0 and  = 2 , we have x = 1.
𝜋 𝑝−1 𝑞−1
1 1
Therefore ∫02 𝑠𝑖𝑛𝑝 𝜃 𝑐𝑜𝑠 𝑞 𝜃 𝑑𝜃 = ∫0 𝑥 2 (1 − 𝑥) 2 . 2 𝑑𝑥

𝑝+1 𝑞+1
1 1
= 2 ∫0 𝑥 ( )−1
2 (1 − 𝑥)( 2
)−1
𝑑𝑥

1 𝑝+1 𝑞+1
= 2B( , ).
2 2

𝑝+1 𝑞+1
Note: By taking = m and = n, we get p = 2m-1 and q = 2n-1.
2 2

𝜋
Therefore B(m, n) = 2 ∫0 𝑠𝑖𝑛2𝑚−1 𝜃 𝑐𝑜𝑠 2𝑛−1 𝜃 𝑑𝜃.
2

3. B(m, n + 1) + B(m + 1, n) = B(m, n).

𝐵(𝑚,𝑛+1) 𝐵(𝑚+1,𝑛) 𝐵(𝑚,𝑛)


4. = =
𝑛 𝑚 𝑚+𝑛

Verification: Left to the student’s practice



Definition: If n is positive then ∫0 𝑒 −𝑥 𝑥 𝑛−1 𝑑𝑥 is called a Gamma function denoted by (n).

Properties of Gamma function:

Property 1: (1) = 1

Verification: By definition (n) = ∫0 𝑒 −𝑥 𝑥 𝑛−1 𝑑𝑥.

Therefore (1) = ∫0 𝑒 −𝑥 𝑥 0 𝑑𝑥

∞ 𝑒 −𝑥 ∞
= ∫0 𝑒 −𝑥 𝑑𝑥 = [ −1 ]
0

= − { lim 𝑒 −𝑥 − 𝑒 0 }
𝑥→∞

= -(0-1) = 1.

Property 2: Reduction formula: (n + 1) = n (n).



Verification: By definition, (n + 1) = ∫0 𝑒 −𝑥 𝑥 𝑛 𝑑𝑥. Applying integration by parts, (n + 1) =
𝑒 −𝑥 ∞ ∞ 𝑒 −𝑥 𝑥𝑛 ∞
[𝑥 𝑛 ( −1 )] - ∫0 −1
𝑛𝑥 𝑛−1 𝑑𝑥 = − { lim 𝑒𝑥
− 0} + 𝑛 ∫0 𝑒 −𝑥 𝑥 𝑛−1 𝑑𝑥
0 𝑥→∞
= -(0-0) + n(n) = n (n).

Property 3: If n is a positive integer then (n) = (n-1)

Verification: (n + 1) = n  (n). Changing n to n-1, we get (n) = (n-1)(n-1). Similarly

(n -1) = (n -2)(n-2).

By substitution (n) = (n-1) (n-2)(n-2).

Similarly, (n - 2) = (n -3)(n-3), and again by substitution, (n) = (n-1)(n-2)(n-3)(n-3).


This process can be continued successively, and it ends with (1), since n is a positive
integer.

Therefore (n) = (n-1) (n-2)(n-3) … 1(1). But (1) = 1.

Hence (n) = (n-1) (n-2)(n-3) … 1.

Property 4: Relation between beta and gamma function:

Γ(𝑚)Γ(𝑛)
𝐵(𝑚, 𝑛) = Γ(𝑚+𝑛)
.


Verification: By definition, ∫0 𝑒 −𝑡 𝑡 𝑛−1 𝑑𝑡, using t as the variable.

Put t = x2. Then dt = 2x dx.

When t = 0, we have x = 0 and when t = , we have x = .


∞ 2 ∞ 2
Therefore (n) = ∫0 𝑒 −𝑥 (𝑥 2 )𝑛−1 2𝑥𝑑𝑥 = 2 ∫0 𝑒 −𝑥 𝑥 2𝑛−1 𝑑𝑥.
∞ 2
Similarly (m) = 2 ∫0 𝑒 −𝑦 𝑦 2𝑛−1 𝑑𝑦.
∞ 2 ∞ 2
Multiplying, (m)  (n) = 2 ∫0 𝑒 −𝑥 𝑥 2𝑛−1 𝑑𝑥  2 ∫0 𝑒 −𝑦 𝑦 2𝑛−1 𝑑𝑦
∞ ∞ 2 +𝑦 2 )
= 4 ∫0 ∫0 𝑒 −(𝑥 𝑥 2𝑛−1 𝑦 2𝑛−1 𝑑𝑥𝑑𝑦 

The range of this integral is the entire first quadrant of the XOY plane. Converting to polar
coordinates, by writing x = r cos , y = r sin  so that dx dy = r d dr.
𝜋
The limits being  from 0 to 2 and r from  to .
𝜋
∞ 2
Therefore (m)  (n) = 4 ∫02 ∫𝑟=0 𝑒 −𝑟 (𝑟 cos 𝜃)2𝑛−1 𝑟 𝑑𝜃 𝑑𝑟 
𝜋
∞ ∞ 2
= 4 (∫02 ∫𝑟=0 𝑠𝑖𝑛2𝑚−1 𝜃 𝑐𝑜𝑠 2𝑛−1 𝜃 𝑑𝜃)  (∫𝑟=0 𝑒 −𝑟 (𝑟)2(𝑚+𝑛)−1 𝑑𝑟)

1 1
= 4(2 𝐵(𝑚, 𝑛)) (2 Γ(𝑚 + 𝑛))

= B(m, n) (m + n).

1 −1
Property 5: (2) = √𝜋 ; ( 2 ) = −2√𝜋.

Γ(𝑚)Γ(𝑛) 1
Verification: Use 𝐵(𝑚, 𝑛) = , by taking 𝑚 = 𝑛 = , we get
Γ(𝑚+𝑛) 2

1 1
1 1 Γ( )Γ( )
𝐵 (2 , 2) = 2 2
. But (1) = 1.
Γ(1)

𝜋 1 1
1 1
Therefore 𝐵 (2 , 2) = 2 ∫02 𝑠𝑖𝑛2(2)−1 𝜃𝑐𝑜𝑠2(2)−1 𝜃𝑑𝜃
𝜋
= 2 ∫02 (sin 𝜃)0 (cos 𝜃)0 𝑑𝜃

𝜋
= 2( − 0) = 𝜋.
2

1
Therefore (2) = √𝜋.

1
Now (n) = (n-1)(n-1)  (n-1) = 𝑛−1 Γ(𝑛).

1 1 1 1
Put 𝑛 = 2, we get Γ (− 2) = 1 Γ(2) = -2√𝜋.

2

1 √𝜋
Property 6: Duplication formula: (m) Γ (𝑚 + 2) = 22𝑚−1 Γ(2𝑚)

Γ(𝑚)Γ(𝑛)
Verification: 𝐵(𝑚, 𝑛) = .
Γ(𝑚+𝑛)

Γ(𝑚)Γ(𝑚) [Γ(𝑚)]2
Therefore 𝐵(𝑚, 𝑚) = = ______(equation 1)
Γ(𝑚+𝑚) Γ(2𝑚)

𝜋⁄
But 𝐵(𝑚, 𝑚) = 2 ∫0 2 𝑠𝑖𝑛2𝑚−1 𝜃 𝑐𝑜𝑠 2𝑚−1 𝜃𝑑𝜃 =
𝜋
2
22𝑚−1
∫02 (2𝑠𝑖𝑛𝜃𝑐𝑜𝑠𝜃)2𝑚−1 𝑑𝜃
𝜋
2
= 22𝑚−1 ∫02 (𝑠𝑖𝑛2𝜃)2𝑚−1 𝑑𝜃. Put 2 = .
1 𝜋
Then d = 2 𝑑𝜃. When  = 0, we have  = 0 and when  = 2 we get  = .

2 𝜋 1
Therefore B(m, m) = 22𝑚−1 ∫0 (𝑠𝑖𝑛𝜙)2𝑚−1 2 𝑑𝜙

𝜋
1
= 22𝑚−1 2 ∫02 (𝑠𝑖𝑛𝜙)2𝑚−1 𝑑𝜙

𝜋 1
1 2𝑚−1 2( )−1
2𝑚−1 2 ∫0 𝑠𝑖𝑛 𝜙𝑐𝑜𝑠 𝜙 𝑑𝜙
= 2 2
2

1
1 1 1 Γ(𝑚).Γ( )
2
= 22𝑚−1 𝐵(𝑚, 2) = 22𝑚−1 1 _______(equation 2).
Γ(𝑚+ )
2

Hence from (1) and (2) we get


1
[Γ(𝑚)]2 1 Γ(𝑚).Γ( ) 1
Γ(2𝑚)
= 1
2
22𝑚−1 Γ(𝑚+ )
. Cancelling (m) both sides and write ( ) = √𝜋 we get
2
2

1 √𝜋
(m) Γ (𝑚 + 2) = 22𝑚−1 Γ(2𝑚).

1 1 𝑛−1
Property 7: (n) = ∫0 (𝑙𝑜𝑔 𝑦) 𝑑𝑦

∞ 1
Hint: Use (n) = ∫0 𝑒 −𝑥 𝑥 𝑛−1 𝑑𝑥. Put e-x = y so that ex = y or x = log .
𝑦

1
Example: Evaluate ∫0 𝑥 4 (1 − 𝑥)3 𝑑𝑥

1
Solution: By definition ∫0 𝑥 𝑚−1 (1 − 𝑥)𝑛−1 𝑑𝑥 = B(m, n).

Take m = p + 1 and n = q + 1.
1
We get ∫0 𝑥 𝑝 (1 − 𝑥)𝑞 𝑑𝑥 = B(p+1, q+1).

1 Γ(5)Γ(4) 4!3! 1
Therefore ∫0 𝑥 4 (1 − 𝑥)3 𝑑𝑥 = B(5, 4) = Γ(5+4)
= 8!
= 280.

𝜋
Example: Evaluate ∫02 𝑠𝑖𝑛5 𝜃 𝑐𝑜𝑠 7 𝜃 𝑑𝜃
𝜋
1 𝑝+1 𝑞+1
Solution: Use ∫02 𝑠𝑖𝑛𝑝 𝜃 𝑐𝑜𝑠 𝑞 𝜃 𝑑𝜃 = 2B( , ).
2 2

𝜋
1 6 8 1 Γ(3)Γ(4) 1 2!3! 1
∫02 𝑠𝑖𝑛5 𝜃 𝑐𝑜𝑠 7 𝜃 𝑑𝜃 = 2B(2 , 2) = 2 Γ(7)
=
2 6!
=
120
.

1
Example: Evaluate ∫0 (𝑥 − 𝑎)𝑝 (𝑏 − 𝑥)𝑞 𝑑𝑥
Solution: Put x = a + (b - a)t.

Then dx = (b - a)dt, x – a = (b - a)t, b – x = (b - a)(1- t).

When x = a, we have t = 0 and when x = b we have t = 1.


1
Therefore ∫0 (𝑥 − 𝑎)𝑝 (𝑏 − 𝑥)𝑞 𝑑𝑥

1
= ∫0 [(𝑏 − 𝑎)𝑡]𝑝 [(𝑏 − 𝑎)(1 − 𝑡)]𝑞 (b-a)

1
= (𝑏 − 𝑎)𝑝+𝑞+1 ∫0 𝑡 𝑝 [(1 − 𝑡)]𝑞 

= (𝑏 − 𝑎)𝑝+𝑞+1 𝐵(p+1, q+1).


𝜋 𝜋
𝑑𝜃
Example: Show that ∫0 √sin 𝜃 𝑑𝜃 ∫0 2 2 𝑑𝜃 = .
√sin 𝜃

𝜋
1 𝑝+1 𝑞+1
Solution: Use ∫0 𝑠𝑖𝑛𝑝 𝜃 𝑐𝑜𝑠 𝑞 𝜃 𝑑𝜃 = 2B(
2 , ), taking p = ½ and q = 0, we get
2 2

𝜋 1 1
1 +1 0+1
∫02 𝑠𝑖𝑛2 𝜃 𝑐𝑜𝑠 0 𝜃 𝑑𝜃 = 2B(2 2 , 2
)

𝜋
1 3 1
 ∫02 √sin 𝜃 𝑑𝜃 = 2B(4 , 2) _______(equation 1)

𝜋 −1 −1
1 +1 0+1
Taking p = -1/2, q = 0, we get ∫02 𝑠𝑖𝑛 2 𝜃 𝑐𝑜𝑠 0 𝜃 𝑑𝜃 = 2B( 2 2 , )
2

𝜋
𝑑𝜃 1 1 1
 ∫02 𝑑𝜃 = 2B(4 , 2) ________(equation 2).
√sin 𝜃

Multiplying (1) and (2)


𝜋 𝜋
𝑑𝜃 1 3 1 1 1 1
∫0 √sin 𝜃 𝑑𝜃 ∫02 √sin 𝜃 𝑑𝜃 = 2B(4 , 2) 2B(4 , 2)
2

2
1 1
1 [Γ (2)] . Γ (4)
=
4 5
Γ( )
4
2 1
1 [√𝜋] .Γ(4)
= 4 5
Γ( )
4

1
1 π.Γ(4)
= 4 1Γ(1)
= .
4 4
∞ 2 √𝜋
Example: Show that ∫0 𝑒 −𝑥 𝑑𝑥 = .
2


Solution: By definition of gamma function (n) = ∫0 𝑒 −𝑡 𝑡 𝑛−1 𝑑𝑡. Put t = x2.

Then dt = 2x dx. When t = 0, we get x = 0; when t =  we get x = .


∞ 2 ∞ 2
Therefore (n) = ∫0 𝑒 −𝑥 (𝑥 2 )𝑛−1 2𝑥𝑑𝑥 = ∫0 𝑒 −𝑥 𝑥 2𝑛−1 𝑑𝑥.

Taking n = 1/2, we get


1
1 ∞ 2 ∞ 2 ∞ 2
(2) = 2∫0 𝑒 −𝑥 𝑥 2(2)−1 𝑑𝑥  √𝜋 = 2∫0 𝑒 −𝑥 𝑥 0 𝑑𝑥 = 2∫0 𝑒 −𝑥 𝑑𝑥.

∞ 2 √𝜋
Therefore ∫0 𝑒 −𝑥 𝑑𝑥 = .
2

𝑎
Example: Evaluate ∫0 𝑥 4 √𝑎2 − 𝑥 2 dx

1
Solution: Put x2 = a2t. Then dx = 𝑎 ∙ 2 𝑡 𝑑𝑡.

When x = 0, we have t = 0; when x = a, we have t = 1.


𝑎
Therefore ∫0 𝑥 4 √𝑎2 − 𝑥 2 dx

1 𝑎
= ∫0 𝑎4 𝑟 2 √𝑎2 − 𝑎2 𝑡 𝑑𝑡
2√𝑡

𝑎6 1 3 1
= 2
∫0 𝑡 2 (1 − 𝑡)2 𝑑𝑡

𝑎6 3 1
= 𝐵 (2 + 1, 2 + 1)
2

𝑎6 5 3
= 𝐵 (2 , 2)
2

5 3
𝑎6 Γ(2Γ(2))
= 2 Γ(4)

31 1
𝑎6 2 2 √𝜋 2 √𝜋
= 2 3!

𝜋𝑎6
= .
32

𝜋
𝜋
Example: Show that ∫02 √tan 𝜃 𝑑𝜃 =
√2
𝜋 𝜋
sin 𝜃
Solution: ∫0 √tan 𝜃 𝑑𝜃 = ∫0 √cos 𝜃 d
2 2

𝜋 1 −1
= ∫02 𝑠𝑖𝑛2 𝜃 𝑐𝑜𝑠 2 𝜃𝑑𝜃
1 −1
1 +1 +1
2 2
=2 𝐵( , )
2 2

1 3 1 3
1 3 1 Γ( )Γ( ) Γ( )Γ( )
4 4 4 4
= 2 B (4 , 4) = 1 3 = .
Γ( + ) Γ(1)
4 4

1
Using Duplication formula with m = 4 , we get

1 3 1 1 1 √𝜋 1
Γ (4) Γ (4) = Γ (4) Γ (4 + 2) = 1 Γ (2) = √2 .
−1
22

𝜋
1 𝜋
Therefore ∫02 √tan 𝜃 𝑑𝜃 = √2 𝜋 = √2.
2

1 𝑑𝑥
Example: Express ∫0 in terms of gamma function.
√(1−𝑥 4 )

−1
1
Solution: Put x2 = sin . Then 𝑑𝑥 = 𝑠𝑖𝑛 2 𝜃 cos 𝜃 𝑑𝜃.
2

1 𝑑𝑥
Therefore ∫0 =
√(1−𝑥 4 )

𝜋 −1
1 𝑠𝑖𝑛 2 𝜃 cos 𝜃 𝑑𝜃
∫0 2
2 √(1−𝑠𝑖𝑛2 𝜃)

𝜋 −1
1
= 2 ∫0 𝑠𝑖𝑛 2 𝜃 𝑑𝜃
2

−1
+1
Γ( 2 )
1 √𝜋 2
=2 −1
2 +2
Γ( 2 )
2

1
√𝜋 Γ(4)
= .
4 Γ(3)
4

Exercises:
1 3
1. Compute (3.5), (4.5), (4) (4)
∞ 2𝑥2
2. Express ∫0 𝑥 𝑛 𝑒 −𝑎 in terms of gamma function.
𝑏
3. Prove that ∫𝑎 (𝑥 − 𝑎)𝑚 (𝑏 − 𝑥)𝑛 𝑑𝑥 = (𝑏 − 𝑎)𝑚+𝑛+1 𝐵(𝑚 + 1, 𝑛 + 1).
1
4. Evaluate ∫0 𝑥 5 (1 − 𝑥 3 )10 𝑑𝑥.
1 𝑥 2 𝑑𝑥 1 𝑑𝑥 𝜋
5. Prove that ∫0 × ∫0 = .
√(1−𝑥 4 ) √(1+𝑥 4 ) 4√2

You might also like