0% found this document useful (0 votes)
50 views3 pages

Gamma Distribution

The document defines the gamma and exponential distributions. The gamma distribution generalizes the exponential distribution and is defined by two parameters, α and λ. When α = 1, the gamma distribution reduces to the exponential distribution. Key properties of the gamma and exponential distributions are derived including expressions for their means, variances, and moment generating functions. Examples are provided to illustrate the exponential distribution in the context of waiting times between events in a Poisson process.

Uploaded by

qwertypushkar123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
50 views3 pages

Gamma Distribution

The document defines the gamma and exponential distributions. The gamma distribution generalizes the exponential distribution and is defined by two parameters, α and λ. When α = 1, the gamma distribution reduces to the exponential distribution. Key properties of the gamma and exponential distributions are derived including expressions for their means, variances, and moment generating functions. Examples are provided to illustrate the exponential distribution in the context of waiting times between events in a Poisson process.

Uploaded by

qwertypushkar123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Gamma and Exponential Distribution

R∞ R1 R∞
Consider the improper integral e−t tα−1 dt = e−t tα−1 dt+ e−t tα−1 dt , where α ∈ R. By
0 0 1
R1 R∞
Limit comparison test, e−t tα−1 dt converges, for all α > 0 and the e−t tα−1 dt converges,
0 1
R∞
for all α ∈ R. Hence, the e−t tα−1 dt is convergent if and only if α > 0.
0

Definition 1. The function Γ : (0, ∞) −→ (0, ∞), defined by,


Z∞
Γ(α) = e−t tα−1 dt
0

is called the gamma function.

Properties:

(1) Γ(α + 1) = αΓ(α), α > 0.


(2) Γ(n) = (n − 1)!, n ∈ N with the convention that 0! = 1.
√ (2n)! √
(3) Γ( 12 ) = π. In general, for n ∈ N ∪ {0}, we have Γ( 2n+1
2
)= n!4n
π.

1. Gamma Distribution

A continuous random variable X is said to have a gamma distribution with parameters


α > 0 and λ > 0 (written as X ∼ G(α, λ) ) if probability density function of X is given
by
( α −λx α−1
λ e x
Γ(α)
, if x > 0
fX (x) =
0, if x ≤ 0

Now, the r-th moment of X ∼ G(α, λ) is


Z∞
E(X r ) = xr fX (x)dx
−∞
Z∞
λα
= xr e−λx xα−1 dx
Γ(α)
0
Z∞
λα
= e−λx x(α+r)−1 dx
Γ(α)
0
Z∞
λα
= e−t t(α+r)−1 dx, (by putting λx = t)
Γ(α)λ(α+r)
0
Γ(α + r)
=
Γ(α)λr
α(α + 1) · · · (α + r − 1)
=
λr
1
Hence

α
E(X) = ;
λ
α(α + 1)
E(X 2 ) = ;
λ2
α
V ar(X) = E(X 2 ) − (E(X))2 = 2 .
λ

The m.g.f. of X ∼ G(α, λ) is

MX (t) = E(etX )
Z∞
= etx fX (x)dx
−∞
Z∞
λα
= e−(λ−t)x xα−1 dx
Γ(α)
0
Z∞
λα
= e−z z α−1 dz, if t < λ (by putting λ − t = z)
Γ(α)(λ − t)α
0
 α
λ
= , if t < λ.
λ−t

Remark 2. Let X ∼ G(α, λ) and h : R −→ R be a function defined by h(x) = λx. Since


the support of X is EX = (0, ∞), the support of Z = h(X) = λX is EZ = (0, ∞). Clearly,
h is strictly increasing on EX . Therefore, the p.d.f. of Z = λX is

(
fX (h−1 (z)) dzd −1
h (z) , if z > 0
fZ (z) =
0, if z ≤ 0
( −z α−1
e z
Γ(α)
, if z > 0
=
0, if z ≤ 0

Hence Z ∼ G(α, 1).

2. Exponential Distribution

A G(1, λ) distribution is called an exponential distribution with parameter λ > 0 and


it is denoted by Exp(λ). Thus p.d.f. of Exp(λ) is

(
λe−λx , if x > 0
f (x) =
0, if x ≤ 0
2
If X ∼ Exp(λ), then
r!)
E(X r ) = ;
λr
1
E(X) = ;
λ
2
E(X 2 ) = 2 ;
λ
1
V ar(X) = E(X 2 ) − (E(X))2 = ;
λ2
λ
MX (t) = , if t < λ.
λ−t
The d.f. of X ∼ Exp(λ) is
Zx
FX (x) = fX (t)dt
−∞
Rx
 λe−λt , if x > 0
= 0
0, if x ≤ 0

(
1 − e−λx , if x > 0
=
0, if x ≤ 0
Remark 3. (1) A Poisson Process is a model for a series of discrete event where the
average time between events is known, but the exact timing of events is random.
(2) The exponential distribution occurs naturally if we consider the distribution of the
length of intervals between successive events in a Poisson process or, equivalently,
the distribution of the interval (i.e. the waiting time) before the first event.
Example 4. The waiting time for occurrence of an event E (say repair time of a machine)
is exponentially distributed with mean of 30 minutes. Find the conditional probability that
the waiting time for occurrence of event E is at least 5 hours given that it has not occurred
in the first 3 hours.

Solution: Let X be the waiting time (in hours) for the occurrence of event E. Then
X ∼ Exp(2). Hence, the required probability is P ({X > 5}|{X > 3}) = PP ({X>5})
({X>3})
=
e−10
e−6
= e−4 .

You might also like