Common Probability Distributions: 1.1 Bernoulli Distribution

G14FOS University of Nottingham
Common probability distributions
1 Discrete distributions
1.1 Bernoulli distribution
Probability model Let X be a random variable which can only assume

two outcomes; 0 if the outcome fo the experiment is a “failure”, 1 if the
outcome of the experiment is a “success”. Furthermore, let p denote the
probability of a success. Hence 1 − p denotes the probability of a failure.
Probability function
px (1 − p)1−x if x = 0, 1

p(x) =
0 otherwise.
Notation X ∼ Ber(p)
Properties
1. E(X) = p
2. Var(X) = p(1 − p)
3. MX (t) = 1 − p + p expt for all real t.
1.2 Binomial distribution
Probability model Suppose an experiment has only two possible out-

comes; “success” or “failure” with probability p and 1 − p, respectively.
Furthermore assume that the experiment is repeated n times (n trials) in
such a way that each trial is independent and the probability of success
remains the same for each trial. Define the random variable X to be the
number of successes observed in the n trials. Then X is said to have a
binomial distribution.
Probability function

 n x
p (1 − p)n−x if x = 0, 1, . . . , n
p(x) = x
0 otherwise.

1
Notation X ∼ Bin(n, p)
Properties
1. E(X) = np
2. Var(X) = np(1 − p)
3. MX (t) = (1 − p + p expt )n for all real t.
4. Let Xi ∼ Bin(ni , p) be independent random variables for i = 1, . . . , k
then
k k
!
X X
Xi ∼ Bin ni , p .
i=1 i=1
Alternative definition Let X1 , X2 , . . . , Xn be a random sample from

of Bernoulli random variables each with probability of success p, then
P n
i=1 Xi ∼ Bin(n, p).
1.3 Poisson distribution
Definition A discrete random variable X which can take on one of the

values 0, 1, 2, . . . is said to be a Poisson random variable if its probability
function is viven by

 e−λ λx
p(x) = if x = 0, 1, 2, . . . for some λ > 0
 x! 0 otherwise.
Notation X ∼ Poi(λ)
Remark The Poisson model was originally developed by Simeon-Denis

Poisson (1837) as an approximation to the binomial distribution when the
number of trials n was large and the probability of success p was small.
It wasn’t until 1889 that the German-Russian mathematician Bortkiewicz
showed that the Poisson model itself was a probability distribution.
Probability model Since the Poisson distribution arose out of the search
for an approximation for the binomial distribution when n is large and p
is small, one can use the Poisson random variable to model experiments
where one wants to count the number of occurrences of an event over
some interval (e.g. time, area). This arises since one can partition the
interval into n small subintervals, where n is a fairly large number and
the probability of success, p, over any one of those subintervals is small.
Then the number of occurrences over the entire interval is approximately
binomial, or Poisson. For example, the number of accidents that occur at
2
an intersection per hour, the number of misprints in a document, or the
number of particles emitted by a radioactive substance in one second can
be modeled by a Poisson random variable.
Properties
1. E(X) = λ
2. Var(X) = λ
t
3. MX (t) = eλ(e −1) for all real t.
4. Let Xi ∼ Poi(λi ) be independent random variables for i = 1, . . . , k
then
k k
!
X X
Xi ∼ Poi λi .
i=1 i=1
2 Continuous distributions
2.1 Uniform distribution
Definition A random variable X is said to be uniformly distributed over

an interval (a, b) if its probability density function is given by
( 1
if a < x < b
f (x) = b−a
0 otherwise.
Notation X ∼ U (a, b)
Properties
a+b
1. E(X) =
2
(b − a)2
2. Var(X) =
 12
 etb − eta
if t 6= 0
3. MX (t) = t(b − a)
1 if t = 0.

3
2.2 Normal distribution
Definition A random variable X is said to have a normal distribution with

parameters µ and σ 2 if its probability density function is given by
(x − µ)2
1 −
f (x) = √ e 2σ 2
2πσ 2
for −∞ < x < ∞.
Notation X ∼ N (µ, σ 2 )
Properties
1. E(X) = µ
2. Var(X) = σ 2
2 2
3. MX (t) = eµt+σ t /2 for all real t.
4. If X ∼ N (µ, σ 2 ) and Y = aX + b for any constants a, b (a 6= 0), then
Y ∼ N (aµ + b, a2 σ 2 ).
5. If X1 , X2 , . . . , Xk are independent random variables where Xi ∼ N (µi , σi2 )
for i = 1, . . . , k then
k k k
!
X X X
2
Xi ∼ N µi , σi .
i=1 i=1 i=1
2.3 Exponential distribution
Definition A random variable X is said to have an exponetial distribution

with parameter λ > 0 if its probability density function is given by
−λx
λe if x ≥ 0
f (x) =
0 otherwise.
Notation X ∼ Exp(λ)
Properties
1
1. E(X) =
λ
1
2. Var(X) = 2
λ
λ
3. MX (t) = if t < λ.
λ−t
4
3 Sampling distributions
3.1 Chi-squared distribution
Definition 1 Let Z be a standard normal random variable, i.e. Z ∼

N (0, 1). Define Y = Z 2 , then Y is said to have a chi-squared distribution
with 1 degree of freedom.
Notation Y ∼ χ21
DefinitionP2 Let Y1 , Y2 , . . . , Yn be independent χ21 random variables. De-

fine X = ni=1 Yi , then X is said to have a chi-squared distribution with
n degrees of freedom.
Notation X ∼ χ2n
Remark An alternative to definition 2 is suppose that Z1 , Z2 , . . . , Zn are

independent standard normal random variables then X = Z12 +Z22 +· · ·+Zn2
is a chi-squared random variable with n degrees of freedom.
Probability density function

1 n n
f (x) = x 2 −1 e− 2
2n/2 Γ( n2 )
if x ≥ 0, where n 1
Z ∞
n x
Γ = x 2 −1 e− 2 dx.
2 2n/2 0
Properties
1. E(X) = n
2. Var(X) = 2n
3. MX (t) = (1 − 2t)−n/2 if t < 21 .
4. If U and V are independent random variables such that U ∼ χ2n and
V ∼ χ2m , then Y = U + V ∼ χ2n+m .
3.2 Student’s t distribution
Definition Let Z be a standard normal random variable, i.e. Z ∼ N (0, 1),

and U be an independent chi-squared random variable with n degrees of
freedom. Define T = √Z , then T is said to have a t distribution with n
U/n
degrees of freedom.
5
Notation T ∼ tn

− n+1
Γ n+1

2 t2 2
f (t) = √ 1 +
nπΓ n2

n
for −∞ < t < ∞.
Properties
1. E(T ) = 0 if n > 1. (If n = 1, E(T ) does not exist.)
n
2. Var(T ) = n−2 if n > 2.
3. MT (t) does not exist since not all moments are finite.
3.3 F distribution
Definition Let U and V be independent random variables such that U ∼

U/n
χ2n and V ∼ χ2m . Define X = , then X is said to have an F
V /m
distribution with n degrees of freedom in the numerator and m degrees of
freedom in the denominator.
Notation X ∼ Fn,m
Γ n+m
n
2 n 2 n −1 n − n+m
2
f (x) = n
m
x 2 1+ x
Γ 2 Γ 2 m m
for x ≥ 0.
Properties
m
1. E(X) = if m > 2.
m−2
2m2 (n + m − 2)
2. Var(X) = if m > 4.
n(m − 2)2 (m − 4)
3. MX (t) does not exist.
4. If T ∼ tn , then T 2 ∼ F1,n .
5. If X ∼ Fn,m , then X1 ∼ Fm,n .

Common Probability Distributions: 1.1 Bernoulli Distribution

Uploaded by

Copyright:

Available Formats

Common Probability Distributions: 1.1 Bernoulli Distribution

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Common Probability Distributions: 1.1 Bernoulli Distribution

Uploaded by

Copyright:

Available Formats

G14FOS University of Nottingham

Common probability distributions

1.1 Bernoulli distribution

Probability model Let X be a random variable which can only assume

1.2 Binomial distribution

Probability model Suppose an experiment has only two possible out-

Alternative definition Let X1 , X2 , . . . , Xn be a random sample from

1.3 Poisson distribution

Definition A discrete random variable X which can take on one of the

Remark The Poisson model was originally developed by Simeon-Denis

2.1 Uniform distribution

Definition A random variable X is said to be uniformly distributed over

Definition A random variable X is said to have a normal distribution with

2.3 Exponential distribution

Definition A random variable X is said to have an exponetial distribution

3.1 Chi-squared distribution

Definition 1 Let Z be a standard normal random variable, i.e. Z ∼

DefinitionP2 Let Y1 , Y2 , . . . , Yn be independent χ21 random variables. De-

Remark An alternative to definition 2 is suppose that Z1 , Z2 , . . . , Zn are

Probability density function

3.2 Student’s t distribution

Definition Let Z be a standard normal random variable, i.e. Z ∼ N (0, 1),

Probability density function

for −∞ < t < ∞.

Definition Let U and V be independent random variables such that U ∼

Probability density function

You might also like