MATH F113: Probability & Statistics

Anupama Sharma

Department of Mathematics,
BITS PILANI K K Birla Goa Campus, Goa

Semester I, 2023-24
Gamma distribution

Gamma function: For ↵ > 0, the gamma function (↵) is defined

as
Z1
(↵) = x ↵ 1 e x dx. (1)
0

Some important properties of functions:

1. (1) = 1.

2. For any ↵ 1, (↵) = (↵ 1) (↵ 1).

3. For any positive integer, n, (n) = (n 1)!

p
4. (1/2) = ⇡.
Gamma distribution

Definition: A continuous random variable X is said to have

gamma distribution if the pdf of X is
(
1 ↵ 1 e x/
(↵) ↵ x x 0
f (x; ↵, ) = .
0, otherwise

where the parameters ↵ and satisfy ↵ > 0, > 0.

Gamma distribution

Definition: A continuous random variable X is said to have

gamma distribution if the pdf of X is
(
1 ↵ 1 e x/
(↵) ↵ x x 0
f (x; ↵, ) = .
0, otherwise

where the parameters ↵ and satisfy ↵ > 0, > 0. The standard

gamma distribution has = 1, so the pdf of a standard gamma rv
is given by,
(
1 ↵ 1e x
(↵) x x 0
f (x; ↵) = .
0, otherwise

If X follows gamma distribution with parameters ↵ and , we write

X ⇠ (↵, ).
Gamma distribution
Gamma distribution
Theorem: Let X be a gamma random variable with parameters ↵
and . Then,
mX (t) = (1 t) ↵ with t < 1/ ,
E [X ] = ↵ , Var [X ] = ↵ 2.
Gamma distribution
Gamma distribution
Gamma distribution
Definition: When X is a standard gamma r.v., the cdf of X ,
Z x ↵ 1 y
y e
F (x; ↵) = dy x >0
0 (↵)

is called Incomplete gamma distribution.

Gamma distribution
Definition: When X is a standard gamma r.v., the cdf of X ,
Z x ↵ 1 y
y e
F (x; ↵) = dy x >0
0 (↵)

is called Incomplete gamma distribution.

Example: Let a X r.v. follows standard gamma distribution with

↵ = 2, then

P(3  X  5) = F (5; 2) F (3; 2)

Gamma distribution
Gamma distribution

Proposition: Let X have a gamma distribution with parameter ↵

and , then for any x > 0 the cdf of X is given by,
x
P(X  x) = F (x; ↵, ) = F ;↵

where F ( · ; ↵) is incomplete gamma distribution.

Gamma distribution
Example: Suppose that when a transistor of a certain type is
subjected to an accelerated life test, the lifetime X (in weeks) has
a gamma distribution with mean 24 weeks and standard deviation
12 weeks. What is the probability that a transistor will last
between 12 and 24 weeks? What is the probability that a transistor
will last at most 24 weeks?
Gamma distribution
Exercise: Prove that
1. Let X1 ⇠ (↵1 , ) and X2 ⇠ (↵2 , ). X1 , X2 are
independent. Then,

X1 + X2 ⇠ (↵1 + ↵2 , ).

2. Let X ⇠ (↵, ). Let k > 0 and Y = kX . Then,

Y ⇠ (↵, k ).
Exponential distribution
distribution with ↵ = 1 and = 1/
Exponential distribution
distribution with ↵ = 1 and = 1/
Definition: A random variable X is said to have exponential
distribution with parameter ( > 0) if the pdf of X is
(
e x for x 0,
f (x; ) =
0 otherwise.
Exponential distribution
distribution with ↵ = 1 and = 1/
Definition: A random variable X is said to have exponential
distribution with parameter ( > 0) if the pdf of X is
(
e x for x 0,
f (x; ) =
0 otherwise.
Exponential distribution

Theorem: Let X be a gamma random variable with parameters ↵

and . Then,
mX (t) = (1 t/ ) 1 with t< ,
1 1
E [X ] = , Var [X ] = 2 .

Theorem: The cdf of F (x; ) of exponential r.v. is given by

(
1 e x for x 0,
P(X  x) = F (x; ) =
0 otherwise.

Note that, Z Z
1 1
x
f (x; ) = e dx = 1
0 0
Exponential distribution

Example: Suppose that the length of a phone call in minutes is an

exponential random variable with parameter = 1/10. If someone
arrives immediately ahead of you at a public telephone booth, find
the probability that you will have to wait
(a) more than 10 minutes;
(b) between 10 and 20 minutes.
Exponential distribution

Proposition: Suppose that the events are occurring according to a

Poisson process with the rate ↵, that is, the number of events
occurring in any time interval of length t has a Poisson distribution
with parameter ↵t (where ↵, the rate of the event process, is the
expected number of events occurring in 1 unit of time) and that
numbers of occurrences in non-overlapping intervals are
independent of one another. Then the distribution of elapsed time
between the occurrence of two successive events is exponential
with parameter = ↵.
In other words, consider a Poisson process with parameter ↵. Let
W denote time of occurrence of first event. Then W has an
exponential distribution with parameter = ↵.
Exponential distribution

Proof: Consider cdf of W .

F (w ; ) = P(W  w ) = 1 P(W > w ).

First occurrence of the event will take place after time w only if no
occurrences of the event in [0, w ]. Let X denote the number of
occurrences of the event in this time. Then, X ⇠ Poisson(↵w ).
Hence,
↵w
P(W > w ) = P(X = 0) = e .
So, we get F (w ) = 1 e ↵w and f (w ) = ↵e ↵w with w 0,
which is the pdf of the exponential distribution with parameter
= ↵.
Exponential distribution
Example: Passengers arrive at an airport according to an
approximate Poisson process at a mean rate of 30 passengers per
hour. What is the probability that the security guard has to wait
more than 3 minutes to check the ID card of the next passenger?
Exponential distribution

Exercise: Let X1 , X2 , . . . Xk be k independent variable such that

each Xi has an exponential distribution with parameter . What is
P k
distribution of Y = Xi ?
i=1
1
Ans: Y ⇠ with parameters k and .

Lemma: Consider a Poisson process with parameter . Let T

denote time of occurrence of k-th event. Then,
✓ ◆
1
T ⇠ ↵ = k, = .