Scribd
Upload
43 views24 pages

Chapter 06

This document provides notes on some common discrete probability distributions: uniform, binomial, multinomial, hypergeometric, negative binomial, geometric, and Poisson. It defines each distribution and gives examples of calculating probabilities and distribution properties like mean and variance. The last sections discuss approximating the hypergeometric with binomial and binomial with Poisson distributions under certain conditions.

Uploaded by

Yamac xx
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
43 views24 pages

Chapter 06

This document provides notes on some common discrete probability distributions: uniform, binomial, multinomial, hypergeometric, negative binomial, geometric, and Poisson. It defines each distribution and gives examples of calculating probabilities and distribution properties like mean and variance. The last sections discuss approximating the hypergeometric with binomial and binomial with Poisson distributions under certain conditions.

Uploaded by

Yamac xx
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/ 24

STA249 Probability and Statistics

Chapter 6 Notes
Some Discrete Probability Distributions

Assoc.Prof.Dr. Özlem KAYMAZ


Ankara University, Faculty of Science, Department of Statistics

E-mail:[email protected]
STA249 Probability and Statistics

Reference Book

This lecture notes are prepared according to the contents of

“PROBABILITY & STATISTICS FOR ENGINEERS & SCIENTISTS by Walpole, Myers, Myers
and Ye”

and

“STA250 Probability and Statistics Course Notes”, written by Assoc.Prof.Dr. Fatih Tank
Discrete Probability Distributions

 The probability distribution for a discrete variable


𝑋 can be represented by a formal, a table, or a graph
that provides 𝑝(𝑥) = 𝑃 (𝑋 = 𝑥).
 Some common discrete distribution models:
 Uniform: All outcomes are equally likely.
 Binomial: Number of successes in n independent
trials, with each trial having probability of success p
and probability of failure q (= 1-p).
 Multinomial: # of outcomes in n trials, with each of k
possible outcomes having probabilities p1, p2, …, pk.
Discrete Probability Distributions

 Common discrete distribution models, continued:


 Hypergeometric: A sample of size n is selected from N
items without replacement, and k items are classified as
successes (N - k are failures).
 Negative Binomial: In n independent trials, with
probability of success p and probability of failure q
(q = 1-p) on each trial, the probability that the kth
success occurs on the xth trial.
 Geometric: Special case of the negative binomial. The
probability that the 1st success occurs on the xth trial.
 Poisson: If  is the rate of occurrence of an event
(number of outcomes per unit time), the probability
that x outcomes occur in a time interval of length t.
Discrete Uniform Distribution

 When X assumes the values x1, x2, …, xk and each


outcome is equally likely. Then
1
f ( x; k )  , x  x , x ,..., x ,
1 2 k
k
and k

x
 i 1 i

 )
k

 (x
2

 
2 i 1 i

k
 Since all observations are equally likely, this is similar
to the mean and variance of a sample of size k, but note
that we use k rather than k - 1 to calculate variance.
Binomial Distribution

 Binomial: Number of successes in n independent


trials, with each trial having probability of success p
and probability of failure q (= 1-p).
• Each trial is called a Bernoulli trial.
• Experiment consists of n repeated trials.
• Two possible outcomes, called success or failure.
• P(success) = p, constant from trial to trial.
• The repeated trials are independent.
 The number of successes in 𝑛 Bernoulli trials is a called
binomial random variable. The probability distribution
of this discrete random variable called the binomial
distribution and its values will be denoted by 𝑏(𝑥; 𝑛, 𝑝).
Binomial Distribution

 Binomial: If 𝑥 is the number of successes in 𝑛 trials,


each with two outcomes where 𝑝 is the probability of
success and 𝑞 = 1 − 𝑝 is the probability of failure, the
probability distribution of 𝑋 is
 n
b( x; n, p )    p q , x  0,1,2,..., n.
x n x

 x
the number of ways a given outcome 𝑥 can occur times
the probability of that outcome occurring, and

  np

  npq
2
Binomial Distribution
Example 1.
Binomial Distribution
Example 2.
Binomial Distribution
Example 3.

Find the mean and variance of the binomial random variable of Example 5.2.
Hypergeometric Distribution

 Hypergeometric: The distribution of the number of


successes, x, in a sample of size n is selected from N
items without replacement, where k items are classified
as successes (and N - k as failures), is
 k  N  k 
  
 x  n  x 
h( x; N , n, k )  , max{0, n  ( N  k )}  x  min{k , n}
N
 
n 
then nk

N
and
N n k k
2
   n  (1  )
N 1 N N
Hypergeometric Distribution
Example
Binomial Approximation to Hypergeometric

 If
n is small compared with N, then the hypergeometric
distribution can be approximated using the binomial
distribution.
 The rule of thumb is that this is valid if (n/N)  0.05. In
this case, we can use the binomial distribution with
parameters n and p = k/N.
 Then

nk k k
  np    npq  n  (1  )
2

N N N
Negative Binomial Distribution

 Negative Binomial: In n independent trials, with


probability of success p and probability of failure q
(q = 1-p) on each trial, the probability that the kth
success occurs on the xth trial.

 x  1
b * ( x; k , p )    p q , x  k , k  1, k  2,...
k xk

 k  1

 Again we have the number of ways an outcome x can


occur times the probability of that outcome occurring.
 The above formula comes from the fact that in order to
get the kth success on the xth trial, we must have
k - 1 successes in the first x - 1 trials, and then the final
trial must also be a success.
Negative Binomial Distribution
Example
Geometric Distribution

 Geometric: Special case of the negative binomial with


k = 1. The probability that the 1st success occurs on the
xth trial is
g ( x; p )  pq , x  1,2,3,...
x 1

1
then 
p

and 1 p
 
2

p 2
Geometric Distribution
Example
Poisson Distribution

 Poisson distribution: If  is the average # of outcomes


per unit time (arrival rate), the Poisson distribution
gives the probability that x outcomes occur in a given
time interval of length t.
A Poisson process has the following properties:
• Memoryless: the number of occurrences in one time interval is
independent of the number in any other disjoint time interval.
• The probability that a single outcome will occur during a very
short time interval is proportional to the size of the time
interval and independent of other intervals.
• The probability that more than one outcome will occur in a
very short time interval is negligible.
 Note
that the rate could be per unit length, area, or
volume, rather than time.
Poisson Distribution

 Poisson distribution: If  is the rate of occurrence of an


event (average # of outcomes per unit time), the
probability that x outcomes occur in a time interval of
length t is

e (t )
 t x

p( x; t )  , x  0,1,2,...
x!

then
    t
2
Poisson Distribution
Example 1

Source: Mathematical Statistics with Applications, Seventh Edition


Dennis D. Wackerly, William Mendenhall III, Richard L. Scheaffer
Poisson Distribution
Example 2
Poisson Approximation to Binomial

 For
a set of Bernoulli trials with n very large and p
small, the Poisson distribution with mean np can be
used to approximate the binomial distribution.
• Needed since binomial tables only go up to n = 20.
 The rule of thumb is that this approximation is valid if
n  20 and p  0.05. (If n  100, the approximation is
excellent if np  10). In this case, we can use the Poisson
distribution with

    np
2

• A different approximation for the binomial can be used for


large n if p is not small.
Poisson Approximation to Binomial
Example
Next Lesson

 Some Continuous Probability Distributions

See you

You might also like