Discrete Distribution

Degenerated random variable(Discrete r.v.):

A degenerate random variable is a random variable that takes on a single, constant value with
a probability of 1. It represents a "point mass" distribution where there is zero variance, as all
probability is concentrated at one point . Formally, for a degenerate random variable the
probability density function is given as :

P(X = x) = 1 ; when x =c Where c is some constant

=0;x≠c

Or

P( X = c) = 1

P(X ≠ c ) = 0
xP(X x) =

I·
Expectation of degenerated random variable:

𝐸 (𝑋) = ∑ 𝑥. 𝑃(𝑋) = 𝑐. 𝑃(𝑋 = 𝑐) + ∑ 𝑥. 𝑃(𝑋 ≠ 𝑐) = 𝑐. 1 + ∑ 𝑥. 0 = 𝑐

𝑥 𝑥≠𝑐 𝑥≠𝑐

Variance of degenerated random varaible:

𝐸 (𝑋 2 ) = ∑ 𝑥 2 . 𝑃 (𝑋) = 𝑐 2 . 𝑃(𝑋 = 𝑐) + ∑ 𝑥 2 . 𝑃(𝑋 ≠ 𝑐) = 𝑐 2 . 1 + ∑ 𝑥 2 . 0 = 𝑐 2

𝑥 𝑥≠𝑐 𝑥≠𝑐

V(X) = E(X 2 ) − [ E(X) ]2 = 𝑐 2 − (𝑐)2 = 𝑐 2 − 𝑐 2 = 0

Cumulative function of degenerated random varaible:

r
F(x) = 0 ; if x < c

= 1 if x ≥ c

yo
#
Mode of the degenerated random varaible:
Since P(X = c) = 1 , hence mode = c
Discrete Uniform Distribution(discrete r.v):
A random variable X is said to follow discrete uniform distribution if its probability mass
function (p.m.f) is given by:
1
P ( X = x) = ; x = 1,2, … , n
n
= 0 ;otherwise

Mean of discrete uniform distribution:

n n
1 1 1 𝑛. (𝑛 + 1)
𝐸 (𝑋) = ∑ 𝑥. 𝑃(𝑋 = 𝑥 ) = ∑ x. = . ∑ x = . (1 + 2 + 3+. . 𝑛 ) =
n n n 2. 𝑛
𝑥 x=1 x=1

(𝑛 + 1)
𝐸(𝑋) =
2
Variance of discrete uniform distribution:
n n
1 1 1
E( X 2 ) = ∑ 𝑥 2 . 𝑃(𝑋 = 𝑥) = ∑ 𝑥 2. = . ∑ 𝑥 2 = . (12 + 22 + 32 +. . 𝑛 2 )
n n n
𝑥 x=1 x=1
𝑛. (𝑛 + 1). (2𝑛 + 1) (𝑛 + 1). (2𝑛 + 1)
= =
6. 𝑛 6
2
(𝑛 + 1). (2𝑛 + 1) (𝑛 + 1)
V( X ) = E( X 2 ) − [ E( X ) ]2 = −[ ]
6 2
(𝑛 + 1). (2𝑛 + 1) (𝑛 + 1)2 (𝑛 + 1) (2𝑛 + 1) (𝑛 + 1)
= − = .[ − ]
6 4 2 3 2

(𝑛 + 1) 2. (2𝑛 + 1) − 3. (𝑛 + 1)
= .[ ]
2 2𝑥3

(𝑛 + 1) 2. (2𝑛 + 1) − 3. (𝑛 + 1)
= .[ ]
2 6

(𝑛 + 1) 4𝑛 + 2 − 3𝑛 − 3
= .[ ]
2 6
(𝑛 + 1) 𝑛 − 1
= .[ ]
2 6
𝑛2 − 1
𝑉(𝑋) =
12
If x = 0,1,2,...,n then the probability mass function is as follows:
1
P ( X = x) = ; x = 0,1,2, … , n
n+1
=0 ;otherwise
Mean of discrete uniform distribution:
n n
1 1 1
𝐸 (𝑋) = ∑ 𝑥. 𝑃(𝑋 = 𝑥 ) = ∑ x. = .∑x = . (0 + 1 + 2 + 3+. . 𝑛 )
n+1 n+1 n+1
𝑥 x=0 x=0
𝑛. (𝑛 + 1) 𝑛
= =
2. (𝑛 + 1) 2
n n
1 1 1
E( X 2 ) = ∑ 𝑥 2 . 𝑃(𝑋 = 𝑥) = ∑ 𝑥 2. = . ∑ 𝑥2 = . (02 + 12 + 22 +. . 𝑛 2 )
n+1 n+1 n+1
𝑥 x=1 x=1
𝑛. (𝑛 + 1). (2𝑛 + 1) 𝑛. (2𝑛 + 1)
= =
6. (𝑛 + 1) 6

𝑛. (2𝑛 + 1) 𝑛 2 2𝑛 2 + 𝑛 𝑛 2 4. (2𝑛 2 + 𝑛) − (6. 𝑛 2 )

𝑉 (𝑋 ) = 𝐸 (X 2 ) − [ E( X ) ]2 = −[ ] = − =
6 2 6 4 24
8𝑛 2 + 4𝑛 − 6𝑛 2 2𝑛 2 + 4𝑛
= =
24 24
2𝑛(𝑛 + 2)
=
24
𝑛(𝑛 + 2)
𝑉(𝑋) =
12
Bernoulli Distribution(discrete r.v):
A Bernoulli trial can be described as follows:
[Link] trial can result in two possible exhaustive and mutually exclusive outcomes termed as
success or failure,yes or no,head or tail,defective or non-defective,etc.
[Link] probability of success ‘p’ is same for each trial.

The probabilities of success and failure can be summarised as Bernoulli distribution.

A discrete random variable X is said to follow Bernoulli Distribution if its probability mass
function is given as :

P(X = x) = px . (1 − p)1−x ; 0 ≤ p ≤ 1, x = 0,1

=0 ; otherwise
Or
x 1−x
P(X = x) = p . q ; 0 ≤ p ≤ 1, x = 0,1 where 1 − p = q
=0 ; otherwise
Eg: [Link] failing or passing exam.

[Link] heads or tail.

[Link] a lottery or not wining a lottery.
Mean of Bernoulli Distribution:
1 1

E(X) = ∑ 𝑥. 𝑃(𝑋 = 𝑥) = ∑ x. P(X = x) = ∑ x. px. (1 − p)1−x

𝑥 x=0 x=0

= 0. p0 . (1 − p)1−0 + 1. p1 . (1 − p)1−1

E(X) = p
Variance of Bernoulli Distribution:
1 1
2)
E( X = ∑ 𝑥 . 𝑃(𝑋 = 𝑥) = ∑ x . P(X = x) = ∑ x 2 . px. (1 − p)1−x
2 2

𝑥 x=0 x=0

= ( 02 . p0 . (1 − p)1−0 ) + ( 12 . p1 . (1 − p)1−1 )

E(X 2 ) = p

V(X) = E(X 2 ) − [ E(X) ]2 = p − p2 = p. (1 − p) = pq

Eg: If a student has 70% chance of passing then X that denotes passing of exam by that
student will follow bernoulli distribution. X=1 if the student passes the exam. X= 0 if the
students fail the exam and here p = 0.7.

Binomial Distribution(discrete r.v):

A Binomial trial can be described as follows:
[Link] trial can result in only two exhaustive and mutually exclusive outcomes termed as
success or failure.

[Link] total number of trials, n, is finite and the trials are independent of each other.

[Link] probability of any outcome of the trail remains fixed over time.

A discrete random variable X is said to follow binomial Distribution if its probability mass
function is given as :

P(X = x) = nCx . px. (1 − p)n−x ; 0 ≤ p ≤ 1, x = 0,1, . . . , n

=0 ; otherwise
Or
P(X = x) = nCx . px. q1−x ; 0 ≤ p ≤ 1, x = 0,1, . . . , n where 1 − p = q
=0 ; otherwise

Eg: [Link] a coin n times and noting the number of heads .

[Link]. of time we hit a target in n shots.

Mean of Binomial Distribution:
n n

E(X) = ∑ x. P(X = x) = ∑ x. nCx . px . (1 − p)n−x

x=0 x=0
n
n!
= ∑ x. . px. (1 − p)n−x
x! (n − x)!
x=1
n
n. (n − 1)!
= ∑ x. . px−1+1 . (1 − p) (n−1)−(x−1)
x. (x − 1)! [ (n − 1) − (x − 1) ]!
x=1
n
(n − 1)!
= n∑ p1 . px−1 . (1 − p)(n−1)−(x−1)
[ ( ) ( )
(x − 1)! n − 1 − x − 1 !]
x=1
n
(n − 1)!
= np ∑ . px−1 . (1 − p) (n−1)−(x−1)
(x − 1)! [ (n − 1) − (x − 1) ]!
x=1

Let y = x-1 so when x=1 y = 0 and when x = n then y = n – 1

there
Yu Binomial (n-1 , p)
Ef
n−1
(n − 1)!

(*j) pY(1-p(3, y ....,

= np ∑ . py . (1 − p)(n−1)−y
>
-

01
y)
[ ( )
y! n − 1 − y !] =

P(y
,
y=0 = =

E(X) = np

Variance of Bernoulli Distribution: :[Plyzy)

=
1 = /
n n

E[ X. (X − 1)] = ∑ x. (x − 1). P(X = x) = ∑ x(x − 1). nCx px . (1 − p) n−x

x=0 x=0
n
n!
= ∑ x. (x − 1) . px . (1 − p) n−x
x! (n − x)!
x=1
n
n. (n − 1). (n − 2)!
= ∑ x. (x − 1) px−2+2 . (1 − p) (n−2)−(x−2)
x. (x − 1). (x − 2)! [ (n − 2) − (x − 2) ]!
x=1
n
(n − 2)!
= n. (n − 1) ∑ p2 . px−2 . (1 − p) (n−2)−(x−2)
(x − 2)! [ (n − 2) − (x − 2) ]!
x=2

n
(n − 2)!
= n. (n − 1). p2 ∑ . px−2 . (1 − p)(n−2)−(x−2)
(x − 2)! [ (n − 2) − (x − 2) ]!
x=2

Let y = x-2 so when x=2 ,y = 0 and when x = n then y = n – 2

 If ~Bin (n-2, p) then
P(y y)= = (4j2)pY(l-p(4-3-2 0 1, ....,
n - 2
where y =
,

n−2
(n − 2)!
:
[P(Y =
y) = 1

>
= n. (n − 1). p2 ∑ . py . (1 − p) (n−2)−y -

y! [ (n − 2) − y ]!
y=0
2
-y
-
n

E [ X. (X − 1)]= n.(n-1). p2

E( X 2 − X )= n.(n-1). p2

E( X 2 ) − E(X ) = n. (n − 1). p

E( X 2 ) = n. (n − 1). p2 + E(X)

E( X 2 ) = n. (n − 1). p2 + n. p

E( X 2 ) = (n2 − n) . p2 + n. p

E( X 2 ) = (n2 p2 ) − n. p2 + n. p

V(X) = E(X 2 ) − [ E(X) ]2 = (n2 p2 ) − n. p2 + n. p − n2 p2 = n. p − n. p2 = np(1 − p)

= npq

Since 0 < p < 1 so p > 1-p multiplyig both side with n as n > 0 then np > np(1-p) i.e.

np > npq . Hence Mean > Variance for binomial distribution.

Recurrence relation for binomial distribution:

Suppose X ~ Binomial (n,p)

n
∴ P(X = x) = Cx px(1 − p) n−x ; x = 0,1,2, . . . n

Consider,
n
P(X = x + 1) Cx+1 . px+1 . (1 − p) n−x−1
= n
P(X = x) Cx . px . (1 − p) n−x

n!
. px+1 . (1 − p) n−x−1
(x + 1)! (n − x − 1)!
=
n!
. px. (1 − p)n−x
x! (n − x)!
n! x! (n − x)!
= . px+1−x. (1 − p)n−x−1−(n−x)
n! (x + 1)! (n − x − 1)!
x! (n − x). (n − x − 1)! 1
= p (1 − p)n−x−1−n+x
(x + 1). x! (n − x − 1)!
(n − x) 1
= . p . (1 − p)−1
(x + 1)
( n − x) . p
=
(x + 1). (1 − p)
 P(X = x + 1) ( n − x) . p
=
P ( X = x) (x + 1). (1 − p)
( n − x) . p
P(X = x + 1) = . P ( X = x)
(x + 1) . (1 − p)

Important features of Binomial Distribution:

1. If X1 , X 2 , … , X n are independent and identically distributed bernoulli variate each with
parameter p, then
n

∑ X i ~Binomial (n, p)
i=1
2. The mode of the distribution is the integral part of (n+1).p if (n+1).p is not an
[Link] (n+1).p is an integer say M then the distribution has two modal values
which are (M-1) and M.
3. If X and Y are independent variable such that X~Bin(n1 , 𝑝) and Y~ Bin(n2 , 𝑝) then
X+Y ~Bin(n1 + n2 , 𝑝). This is known as additive property of Binomial distribution.

Poisson distribution(discrete r.v):

Consider an event whose occurrence is rare but it occurs on average λ times during certain
interval of time. Then the random varaible X denoting number of occurrence of the event
during the interval follows Poisson distribution.

A discrete random variable X is said to follow Poisson distribution if it assumes only non-
negative values an its probability mass function (p.m.f) is given as :

e−λ . λx
P(X = x) = ; x = 0,1,2. . . . . , ∞, λ > 0
x!
=0 ; otherwise

Mean of Poisson Distribution:

∞
e−λ . λx
E(X) = ∑ x. P(X) = ∑ x.
x!
x x=0
∞
e−λ . λx−1+1
= ∑ x.
x. (x − 1)!
x=1
∞
e−λ . λx−1 . λ1
=∑
(x − 1)!
x=1
∞
e−λ . λx−1
= λ. ∑
(x − 1)!
x=1

Let y = x-1 so when x = 1 then y = 0 and when x = ∞ y = ∞

 If Y Poisson (X)

↑
Ply y)
:
=

∞
e−λ . λy
= λ. ∑
(y)! 1
y=0
: [P(y = y) =

.
E(X) = λ
3
Varaince of Poisson Distribution:
∞
e−λ . λx
E[ X. (X − 1) ] = ∑ x. (x − 1). P(X) = ∑ x. (x − 1)
x!
x x=0
∞
e−λ . λx−2+2
= ∑ x. (x − 1).
x. (x − 1). (x − 2)!
x=2

∞
e−λ . λx−2 . λ2
=∑
x=2
(x − 2)!
If Y Poisson (X)

:
Let y = x-2 so when x = 2 then y = 0 and when x = ∞ y = ∞
Ply-yl y
= λ2 . ∑
∞
e−λ . λy
(y)!
&
y=0
1
E[ X. (X − 1) ] = λ2
: [P(y = y) =

.
3
E(X 2 − X) = λ2

E(X 2 ) − E(X) = λ2

E(X 2 ) − λ = λ2

E(X 2 ) = λ2 + λ

V(X) = E(X 2 ) − [ E(X) ]2 = λ2 + λ − λ2 = λ

Reccurence relation for Poisson distribution:

Suppose X ~ Poisson (λ)

e−λ . λx
∴ P(X = x) = ; x = 0,1,2, . . . ∞
x!
Consider,

e−λ . λx+1
P(X = x + 1) (x + 1)! e−λ . x!
= = . λx+1−x
P(X = x) e−λ . λx (x + 1)! e−λ
x!
 x! 1
= . λ1 = .λ
(x + 1). x! x+1
P(X = x + 1) λ
=
P ( X = x) x+1
λ
P(X = x + 1) = . P ( X = x)
x+1
Poisson approximation to binomial distribution:

If n is large i.e. n ⟶ ∞ and p is small i.e. p ⟶ 0 in a binomial distribution such that np = λ

where λ is finite,then binomial dsitribution tends to poisson distribution i.e.

e−λ . λx
lim [ nCx . px . (1 − p)n−x] = where λ = np
n→∞ x!
Proof:

Suppose X ~ Binomial (n,p)

n
∴ P(X = x) = Cx px(1 − p) n−x ; x = 0,1,2, . . . n

Consider,
n!
lim [ nCx . px. (1 − p)n−x] = lim [ . px . (1 − p) n−x]
n→∞ n→∞ x! (n − x)!
λ
As , λ = np so p =
n

n n. (n − 1). (n − 2). . . (n − x + 1). (n − x)! λ x λ n−x

lim [ Cx . px . (1 − p)n−x ] = lim [ . ( ) . (1 − ) ]
n→∞ n→∞ x! (n − x)! n n

n. (n − 1). (n − 2) … (n − x + 1) λ x λ n−x
= lim [ . ( ) . (1 − ) ]
n→∞ x! n n

λx n. (n − 1). (n − 2). . . (n − x + 1). (n − x)! λ n λ −x

= . lim [ . (1 − ) . (1 − ) ]
x! n→∞ nx n n

λx n. (n − 1). (n − 2). . . (n − x + 1). (n − x)! λ n λ −x

= . lim [ ] . lim [(1 − ) ] . lim [(1 − ) ]
x! n→∞ nx n⟶∞ n n⟶∞ n

λx n n−1 n−2 n−x λ n λ −x

= . lim {( ) . ( ).( )...( )} . lim [(1 − ) ] . lim [(1 − ) ]
x! n→∞ n n n n n⟶∞ n n⟶∞ n

λx 1 2 x λ n λ −x
= . lim {1. (1 − ) . (1 − ) . . . (1 − )} . lim [(1 − ) ] . lim [(1 − ) ]
x! n→∞ n n n n⟶∞ n n⟶∞ n

Consider,
 1 2 x 1 2 x
lim {1. (1 − ) . (1 − ) . . . (1 − )} = lim 1 . lim (1 − ) . lim (1 − ) … lim (1 − )
n→∞ n n n n→∞ n→∞ n n→∞ n n→∞ n
1 1 2 x
Now, lim ( ) = 0 , hence lim (1 − ) = lim (1 − ) = . . . = lim (1 − ) = 1
n→∞ n n→∞ n n→∞ n n→∞ n

1 2 x
lim {1. (1 − ) . (1 − ) . . . (1 − )} = 1.1.1. . . .1 = 1 → A
n→∞ n n n

1
Consider, lim ( ) = 0
n→∞ n

λ −x
lim [(1 − ) ] = lim [(1 − 0)−x ] = lim [1−x ] = lim [1] = 1 ⟶ B
n⟶∞ n n⟶∞ n⟶∞ n⟶∞

λ n
lim [(1 − ) ] = e−λ ⟶ C
n⟶∞ n
λx 1 2 x λ n λ −x
lim [ nCx . px. (1 − p)n−x ] = . lim {1. (1 − ) . (1 − ) . . . (1 − )} . lim [(1 − ) ] . lim [(1 − ) ]
n→∞ x! n→∞ n n n n⟶∞ n n⟶∞ n

From A, B and C we can say :

λx
lim [ nCx . px . (1 − p)n−x] = . 1. e−λ . 1
n→∞ x!
λx −λ
lim [ nCx . px. (1 − p)n−x] = .e
n→∞ x!

Note: The thumb rule stating that Poisson distribution is a good approximation of binomial
distribution if n is atlest 20 and p ≤ 0.05 .

Hypergeometric distribution:
The hypergeometric distribution models the probability of obtaining a specific number of
successes in each number of draws from a finite population. Unlike the binomial
distribution, which assumes replacement, the hypergeometric distribution does not replace
items once they are drawn.

Notations:
 Population Size (N): The total number of items in the population.
 Number of Successes in Population (K): The number of items in the population
that are classified as successes.
 Sample Size (n): The number of items drawn from the population.
 Number of Successes in Sample (x): The number of successes observed in the
drawn sample.
The probability of obtaining exactly k successes in n draws from a population of size N
containing K successes is given by the hypergeometric probability mass function:
(𝐾𝑥). (𝑁−𝐾
𝑛−𝑥
)
𝑃 (𝑋 = 𝑥 ) = ; 0 ≤ 𝑥 ≤ 𝑚𝑖𝑛 (𝐾, 𝑛 ), 𝑁, 𝐾, 𝑛 ∈ ℤ+
(𝑁
𝑛
)

=0 ; otherwise
Expectation of Hypergeometric distribution:
min(K,n)
(𝐾𝑥). (𝑁−𝐾
𝑛−𝑥
)
E(X) = ∑ x. P(X) = ∑ x.
(𝑁
𝑛
)
x x=0

𝐾!
min(K,n)
𝑥. . (𝑁−1−𝐾+1)
𝑥! (𝐾 − 𝑥 )! 𝑛−𝑥−1+1
= ∑
𝑁!
x=1
𝑛! (𝑁 − 𝑛 )!
𝐾. (𝐾 − 1)!
𝑚𝑖𝑛(𝐾,𝑛)
𝑥. . ((𝑁−1)−(𝐾−1))
𝑥. (𝑥 − 1)! (𝐾 − 𝑥 − 1 + 1)! (𝑛−1)−(𝑥−1)
= ∑
𝑁. (𝑁 − 1)!
x=1
𝑛. (𝑛 − 1)! (𝑁 − 𝑛 )!
(𝐾 − 1)!
𝐾
𝑚𝑖𝑛(𝐾,𝑛)
. ((𝑁−1)−(𝐾−1))
(𝑥 − 1)! (𝐾 − 𝑥 − 1 + 1)! (𝑛−1)−(𝑥−1)
= . ∑
𝑁 (𝑁 − 1)!
𝑛 x=1
(𝑛 − 1)! (𝑁 − 𝑛 − 1 + 1)!
(𝐾 − 1)!
𝑛. 𝐾
𝑚𝑖𝑛(𝐾,𝑛)
. ((𝑁−1)−(𝐾−1))
(𝑥 − 1)! [ (𝐾 − 1) − (𝑥 − 1) ]! (𝑛−1)−(𝑥−1)
= . ∑
𝑁 (𝑁 − 1)!
x=1
(𝑛 − 1)! [ (𝑁 − 1) − (𝑛 − 1) ]!
𝑚𝑖𝑛(𝐾,𝑛) 𝐾−1
𝑛. 𝐾 ( 𝑥−1 ). ((𝑁−1)−(𝐾−1)
(𝑛−1)−(𝑥−1)
)
= . ∑
𝑁 (𝑁−1
𝑛−1
)
x=1

Let y = x-1 so when x = 1 then y = 0 also the upper most value will be min( K-1 , n-1)

Considerhypergoneaa
𝑚𝑖𝑛(𝐾−1 , 𝑛−1) 𝐾−1
𝑛. 𝐾 ( 𝑦 ) . ((𝑁−1)−(𝐾−1)
(𝑛−1)−𝑦
) >
-

= . ∑
𝑁 (𝑁−1
𝑛−1
)
y=0

Variance of Hypergeometric distribution:

𝐸(𝑋) =
𝑛. 𝐾
𝑁
-PM =
y) =

(
where 1, DK-1, n 1)
y 0 min
-

=
, .,
.
. .

𝑚𝑖𝑛(𝐾,𝑛)
(𝐾𝑥). (𝑁−𝐾 ) /
𝐸[ X. (X − 1) ] = ∑ 𝑥(𝑥 − 1). 𝑃 (𝑋) = ∑ x. (x − 1). 𝑛−𝑥
[Ply y) =
=

(𝑁 )
x x=0 𝑛 Y
𝐾!
𝑚𝑖𝑛(𝐾,𝑛) 𝑥. (𝑥 − 1). . (𝑁−2−𝐾+2)
𝑥. (𝑥 − 1). (𝑥 − 2)! (𝐾 − 𝑥 )! 𝑛−𝑥−2+2
= ∑
𝑁!
x=1
𝑛! (𝑁 − 𝑛 )!
𝐾. (𝐾 − 1). (𝐾 − 2)!
𝑚𝑖𝑛(𝐾,𝑛)
. ((𝑁−2)−(𝐾−2))
(𝑥 − 2)! (𝐾 − 𝑥 − 2 + 2)! (𝑛−2)−(𝑥−2)
= ∑
𝑁. (𝑁 − 1)(𝑁 − 2)!
x=2
𝑛. (𝑛 − 1). (𝑛 − 2)! (𝑁 − 𝑛 − 2 + 2) !
 (𝐾 − 2)!
𝐾. (𝐾 − 1)
𝑚𝑖𝑛(𝐾,𝑛)
. ((𝑁−2)−(𝐾−2))
(𝑥 − 2)! (𝐾 − 𝑥 − 2 + 2)! (𝑛−2)−(𝑥−2)
= . ∑
𝑁. (𝑁 − 1) (𝑁 − 2)!
x=2
𝑛. (𝑛 − 1) (𝑛 − 2)! (𝑁 − 𝑛 − 2 + 2)!
(𝐾 − 2)!
𝑛. (𝑛 − 1). 𝐾. (𝐾 − 1)
𝑚𝑖𝑛(𝐾,𝑛)
. ((𝑁−2)−(𝐾−2))
(𝑥 − 2)! [ (𝐾 − 2) − (𝑥 − 2) ]! (𝑛−2)−(𝑥−2)
= . ∑
𝑁. (𝑁 − 1) (𝑁 − 2)!
x=2
(𝑛 − 2)! [ (𝑁 − 2) − (𝑛 − 2) ]!
𝑚𝑖𝑛(𝐾,𝑛) 𝐾−2
𝑛. (𝑛 − 1)𝐾. (𝐾 − 1) ( 𝑥−2 ). ((𝑁−2)−(𝐾−2)
(𝑛−2)−(𝑥−2)
)
= . ∑
𝑁. (𝑁 − 1) (𝑁−2
𝑛−2
)
x=2

Let y = x - 2 so when x = 2 then y = 0 also the upper most value will be min( K-2 , n-2)
𝑚𝑖𝑛(𝐾−2 , 𝑛−2) 𝐾−2
𝑛. (𝑛 − 1). 𝐾. (𝐾 − 1) ( 𝑦 ) . ((𝑁−2)−(𝐾−2)
(𝑛−2)−𝑦
)
= . ∑

Cusiderhypergoneaa
𝑁. (𝑁 − 1) (𝑁−2
𝑛−2
)
y=0

𝑛. (𝑛 − 1). 𝐾. (𝐾 − 1)
𝐸[ 𝑋. (𝑋 − 1) ] =
𝑁. (𝑁 − 1)

:Please
n. (n − 1). K. (K − 1)
E( X 2 − X) =
N. (N − 1)
n. (n − 1). K. (K − 1)
E( X 2 ) − E(X) =
N. (N − 1) /
[Ply y) =
=

n. (n − 1). K. (K − 1) Y
E( X 2 ) = + E(X)
N. (N − 1)
n. (n − 1). K. (K − 1) 𝑛. 𝐾 𝑛. 𝐾 (n − 1)(K − 1)
E( X 2 ) = + = [ + 1]
N. (N − 1) 𝑁 𝑁 (𝑁 − 1)

𝑛. 𝐾 (n − 1)(K − 1) + (N − 1)
𝐸 (X 2 ) = [ ]
𝑁 (𝑁 − 1)

V(X) = E(X 2 ) − [ E (X) ]2

𝑛. 𝐾 (𝑛 − 1)(𝐾 − 1) + (𝑁 − 1) 𝑛. 𝐾 2
= [ ]−[ ]
𝑁 (𝑁 − 1) 𝑁

𝑛. 𝐾 (𝑛 − 1)(𝐾 − 1) + (𝑁 − 1) 𝑛. 𝐾
= [ − ]
𝑁 (𝑁 − 1) 𝑁

𝑛. 𝐾 𝑁. (𝑛 − 1). (𝐾 − 1) + N. (𝑁 − 1) − 𝑛𝐾. (𝑁 − 1)
= [ ]
𝑁 𝑁(𝑁 − 1)

𝑛. 𝐾 (𝑁𝑛 − N). (𝐾 − 1) + (𝑁 2 − N) − (𝑛𝐾𝑁 − 𝑛𝐾)

= [ ]
𝑁 𝑁(𝑁 − 1)
 𝑛. 𝐾 nNK − nN − NK + N + 𝑁 2 − N − 𝑛𝐾𝑁 + 𝑛𝐾
= [ ]
𝑁 𝑁(𝑁 − 1)

𝑛. 𝐾 −nN − NK + 𝑁 2 + 𝑛𝐾
= [ ]
𝑁 𝑁(𝑁 − 1)

𝑛. 𝐾 𝑁 2 − 𝑁𝐾 + 𝑛𝐾 − 𝑛𝑁
= [ ]
𝑁 𝑁(𝑁 − 1)

𝑛. 𝐾 𝑁(𝑁 − 𝐾) − 𝑛(𝑁 − 𝐾)
= [ ]
𝑁 𝑁(𝑁 − 1)
𝑛. 𝐾 (𝑁 − 𝐾)(𝑁 − 𝑛)
𝑉(𝑋) = .[ ]
𝑁 𝑁 (𝑁 − 1)
Hypergeomtric distribution approximation to binomial distribution:
𝐾
If N is large i.e. N ⟶ ∞ and K⟶ ∞ in a hypergeometric distribution such that = 𝑝 where
𝑁
p is finite,then Hypergeometric dsitribution tends to Binomial distribution i.e.

(𝐾𝑥). (𝑁−𝐾
𝑛−𝑥
) 𝐾
lim = nCx . px . (1 − p)n−x where =p
N,K→∞ (𝑁
𝑛
) 𝑁

𝐾! (𝑁 − 𝐾 )!
(𝐾𝑥). (𝑁−𝐾 ) .(
𝑛−𝑥 𝑥! 𝐾 − 𝑥 ! 𝑛 − 𝑥 ! (𝑁 − 𝐾 − 𝑛 + 𝑥 )!
( ) )
lim = lim
N,K→∞ (𝑁
𝑛
) N,K→∞ 𝑁!
𝑛! 𝑁 − 𝑛 )!
(

𝐾! (𝑁 − 𝐾)! 𝑛! (𝑁 − 𝑛)!
= lim [ . . . ]
𝑁,𝐾→∞ (𝐾 − 𝑥)! (𝑁 − 𝐾 − 𝑛 + 𝑥)! 𝑥! (𝑛 − 𝑥)! 𝑁!

= lim [ --
𝐾. (𝐾 − 1). (𝐾 − 2). . (𝐾 − 𝑥 + 1). (𝐾 − 𝑥)! (𝑁 − 𝐾). (𝑁 − 𝐾 − 1). (𝑁 − 𝐾 − 2). . (𝑁 − 𝐾 − 𝑛 + 𝑥 + 1)(𝑁 − 𝐾 − 𝑛 − 𝑥)!
. .
𝑛!
𝑁,𝐾→∞
-(𝐾 − 𝑥)! - (𝑁 − 𝐾 − 𝑛 + 𝑥)! 𝑥! (𝑛 −𝑥

𝑛! 𝐾. (𝐾 − 1). (𝐾 − 2). . (𝐾 − 𝑥 + 1) . (𝑁 − 𝐾). (𝑁 − 𝐾 − 1). (𝑁 − 𝐾 − 2). . (𝑁 − 𝐾 − 𝑛

= lim [
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ 𝑁. (𝑁 − 1). . (𝑁 − 𝑛 + 1).
𝐾. (𝐾 − 1). (𝐾 − 2). . (𝐾 − 𝑥 + 1) (𝑁 − 𝐾). (𝑁 − 𝐾 − 1). (𝑁 − 𝐾 − 2). . (𝑁 − 𝐾 − 𝑛
𝑛! .
= lim [ 𝑁𝑥 𝑁 𝑛−𝑥
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ 𝑁. (𝑁 − 1). . (𝑁 − 𝑛 + 1)
𝑁𝑛
𝐾 𝐾−1 𝐾−𝑥+1 𝑁−𝐾 𝑁−𝐾−1 𝑁−𝐾−𝑛+𝑥+1
𝑛! ( ).( )..( ).( ).( )..( )
= lim [ 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 ]
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ 𝑁 𝑁−1 𝑁−𝑛+1
( ).( )..( )
𝑁 𝑁 𝑁
𝐾 𝐾 1 𝐾 1−𝑥 𝑁 𝐾 𝑁 𝐾 1 𝑁 𝐾 1+𝑥−𝑛
𝑛! ( ).( − )..( + ).( − ).( − − )..( − + )
= lim [ 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 𝑁 ]
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ 𝑁 𝑁 1 𝑁 1−𝑛
( ).( − )..( + )
𝑁 𝑁 𝑁 𝑁 𝑁
 1 1−𝑥 1 1+𝑥−𝑛
𝑛! (𝑝). (𝑝 − ) . . (𝑝 + ) . (1 − 𝑝) . (1 − 𝑝 − ) . . (1 − 𝑝 + )
= lim [ 𝑁 𝑁 𝑁 𝑁 ]
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ 1 1−𝑛
(1). (1 − ) . . (1 + )
𝑁 𝑁
1
As, lim ( ) = 0
N→∞ N

𝑛! (𝑝). (𝑝 − 0) . . (𝑝 + 0). (1 − 𝑝) . (1 − 𝑝 − 0). . (1 − 𝑝 + 0)

= lim [ ]
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ (1). (1 − 0) . . (1 + 0)

𝑛! 𝑝𝑥 . (1 − 𝑝) 𝑥
= lim [ ]
𝑥! (𝑛 − 𝑥)! 𝑁,𝐾→∞ 1

(𝐾 𝑁−𝐾
𝑥 ).( 𝑛−𝑥 ) 𝑛!
lim = . 𝑝𝑥 . (1 − 𝑝) 𝑥
N,K→∞ (𝑁
𝑛) 𝑥!(𝑛−𝑥)!

Continous Distribution
Exponential Distribution:
The amount of time between occurrence of random events like length of time between
breakdown of an equipment,length of time between two successive arrivals in the queue
(inter-arrival time) are often described using exponential distribution.
A continous random variable X is said to follow exponential distribution if it assumes the
following probability density function:

f(x) = λe−λx ; 0 < x < ∞

= 0 ; otherwise

Cumulative distribution function:

If x< 0 then F(X) = 0

x
x x e−λx e−λ.x e −λ.0
If x > 0 then F(X)= P( X ≤ x) = ∫0 λe−λx dx = λ ∫0 e−λx dx = λ [ ] = λ[ − ]
−λ 0 −λ −λ

e−λ.x 1 e−λ.x − 1 e−λ.x − 1

F(X) = λ [ − ] = λ[ ]= = 1 − e−λ.x
−λ −λ −λ −1

Mean of the exponential distribution:

∞ ∞

E(X) = ∫ x. f(x) dx = ∫ x. λe−λx dx = λ ∫ x. e−λx dx

x 0 0
∞
(a − 1)!
∫ x a−1 . e−bx dx =
ba
0

So here a =2 and b = λ
So
∞
(2 − 1)! 1! 1
∫ x 2−1 . e−λx dx = = =
λ2 λ2 λ2
0
∞
1 1
E(X) = λ ∫ x. e−λx dx = λ. 2
=
λ λ
0

Variance of exponential distribution:

∞ ∞

E(X 2 ) = ∫ x 2 . f(x) dx = ∫ x 2 . λe−λx dx = λ ∫ x 2 . e−λx dx

x 0 0
∞
(a − 1)!
∫ x a−1 . e−bx dx =
ba
0

So here a =3 and b = λ
So
∞
(3 − 1)! 2! 2
∫ x 3−1 . e−λx dx = 3
= 3= 3
λ λ λ
0
∞
2 2
E(X 2 ) = λ ∫ x 2 . e−λx dx = λ. 3
= 2
λ λ
0

2 1 2 2 1 2−1 1
V ( X ) = E( X 2 ) − [ E( X ) ] 2 = − ( ) = − = =
λ2 λ λ2 λ2 λ2 λ2
Median of Exponential Distribution:

If M is the median then P(X ≤ M) = 0.5

P(X ≤ M) = F(M)

We know that,

F(X) = 1 − e−λ.x so, F(M) = 1 − e−λ.M

F(M)=0.5

1 − e−λ.M = 0.5

−e−λ.M = 0.5 − 1

−e−λ.M = −0.5

e−λ.M = 0.5

loge e−λ.M = loge 0.5

 −λ. M. loge e = loge 0.5

−λ. M. 1 = loge 0.5

loge 0.5
M=
−λ
Rectangle or Uniform Distribution(Continous r.v.):
1
f(x) = ; a< x <b
𝑏−𝑎
= 0 ; otherwise

Cumulative distribution function:

If x< a then F(X) = 0
If x > a then,
x x
1 1
𝐹(𝑋) = 𝑃( 𝑋 ≤ 𝑥) = ∫ dx = . ∫ 1. dx
𝑏−𝑎 𝑏−𝑎
a a

1
= . [x]xa
𝑏−𝑎
1
= [x − a]
𝑏−𝑎
𝑥 − a
F(X) =
b − a
Mean of the Uniform distribution:
b b
1 1
E(X) = ∫ x. f(x)dx = ∫ 𝑥. dx = ∫ x dx
𝑏−𝑎 𝑏−𝑎
x a a

𝑏
1 𝑥2 1 𝑏 2 𝑎2
= .[ ] = .[ − ]
𝑏−𝑎 2 𝑎 𝑏−𝑎 2 2

1 𝑏 2 − 𝑎2 1 (𝑏 + 𝑎 ). (𝑏 − 𝑎 ) (𝑏 + 𝑎)
= .[ ]= . =
𝑏−𝑎 2 𝑏−𝑎 2 2

Variance of Uniform distribution:

b b
1 1
E(𝑋 2 ) = ∫ 𝑥 2 . f(x)dx = ∫ 𝑥 2 . dx = ∫ 𝑥 2 dx
𝑏−𝑎 𝑏−𝑎
x a a

𝑏
1 𝑥3 1 𝑏 3 𝑎3
= .[ ] = .[ − ]
𝑏−𝑎 3 𝑎 𝑏−𝑎 3 3
 1 𝑏 3 − 𝑎3
= .[ ]
𝑏−𝑎 3

1 (𝑏 2 + 𝑎𝑏 + 𝑎2 ). (𝑏 − 𝑎) (𝑏 2 + 𝑎𝑏 + 𝑎2 )
= . =
𝑏−𝑎 3 2
(𝑏 2 + 𝑎𝑏 + 𝑎2 ) b+a 2
V ( X ) = E ( X 2 ) − [ E( X ) ] 2 = −( )
3 2
𝑏 2 + 𝑎𝑏 + 𝑎2 𝑏 2 + 2𝑎𝑏 + 𝑎2
= −
3 4
4. (𝑏 2 + 𝑎𝑏 + 𝑎2 ) − 3(𝑏 2 + 2𝑎𝑏 + 𝑎2 )
=
12
4𝑏 2 + 4𝑎𝑏 + 4𝑎2 − 3𝑏 2 − 6𝑎𝑏 − 3𝑎2
=
12
𝑏 2 − 2𝑎𝑏 + 𝑎2 (𝑏 − 𝑎 )2
= =
12 12
Median of Uniform Distribution:

If M is the median then P(X ≤ M) = 0.5

P(X ≤ M) = F(M)

We know that,
𝑥−𝑎 𝑀−𝑎
F(X) = so, F(M) =
𝑏−𝑎 𝑏−𝑎
F(M)=0.5
𝑀−𝑎
= 0.5
𝑏−𝑎
M − a = 0.5(b − a)

M − a = 0.5b − 0.5a

M = 0.5b − 0.5a + a

𝑀 = 0.5𝑏 + 0.5𝑎

𝑀 = 0.5. (𝑏 + 𝑎)
(𝑏 + 𝑎)
𝑀=
2