Discrete Variable Probability Distribution Functions

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 47

Discrete Variable Probability

Distributions
ENGDAT1 Lecture
Eric A. Siy
Dept of Industrial Engineering
De La Salle University

Siy, Discrete Probability Distributions, DLSU Industrial Engineering Dept


Commonly Studied Patterns of Probability in
Discrete Random Variables
• Discrete Uniform distribution
• Binomial Distribution
• Negative Binomial and Geometric Distribution
• Multinomial Distribution
• Hypergeometric Distribution
• Multivariate Hypergeometric Distribution
• Poisson Distribution

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Discrete Uniform Distribution
• Occurs if each outcome in an experiment (or observation) has equal
probability of occurrence.
Ex: Raffle tickets being chosen for a particular prize, given the number
of n raffle tickets still not yet chosen.
Ex: Picking a result from the throw of a die (1, 2, 3, 4, 5 or 6)

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Discrete Uniform Distribution
Parameters: a = Lower limit;
b = upper limit

Probability Distribution Function :

Cumulative Distribution Function:

Mean E(X) = Variance =

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Discrete Uniform Distribution
Most commonly known pattern: throwing a die P(x) = 1/6 for X=1,2,3,4,5,6

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Discrete Uniform Distribution
• There are 75 Bingo chips initially in a Bingo game. The probability of
any one of these chips being chosen for the first draw is:

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Discrete Uniform Distribution
• There are 75 Bingo chips initially in a Bingo game. What is the
probability that the first chip would be from the letter I (#16-30)
P(16<=X<=30) =
=P(X<=30) – P(X<=15)
=
=

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Binomial Distribution Function
• A special type of random variable--called a binomial random
variable--have several characteristics, but the key one is that the
experiment that is being performed has only two possible outcomes -
success or failure. We also call this result a Bernoulli trial.

(Ex) a free throw in basketball - either the player makes a goal or he


doesn't. Another example would be a flipped coin - it's either heads or
tails. A multiple choice test where you're totally guessing would be
another example - each question is either right or wrong.

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Binomial Distribution
Function
A binomial experiment is an experiment which satisfies these four conditions:

• A fixed number of trials


• Each trial is independent of the others
• There are only two outcomes
• The probability of each outcome remains constant from trial to trial.

In short: An experiment with a fixed number of independent trials, each of which can only
have two possible outcomes.

(Since the trials are independent, the probability remains constant.)


Siy, Discrete Probability Distributions, DLSU Industrial
Engineering Dept
Binomial Distribution
Function
A binomial experiment is an experiment which satisfies these four conditions:

• A fixed number of trials


• Each trial is independent of the others
• There are only two outcomes
• The probability of each outcome remains constant from trial to trial.

In short: An experiment with a fixed number of independent trials, each of which can only
have two possible outcomes.

(Since the trials are independent, the probability remains constant.)


Siy, Discrete Probability Distributions, DLSU Industrial
Engineering Dept
Binomial distribution = Free throw distribution

Fixed number of trials = 2


Each trial is independent of the other
There are only two outcomes = in or out
The probability of each trial is constant from
one trial to the next = FT% = 93% for #30

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Binomial Distribution
Parameters: n= number of attempts;
p = probability of “success” per attempt
q = (1-p) = P(failure) per attempt
Random Variable X = number of “successful” results among the n attempts.

Probability Distribution Function :

Cumulative Distribution Function:

Mean E(X) = np Variance =npq


Siy, Discrete Probability Distributions, DLSU Industrial
Engineering Dept
In the astrology zodiac, exactly 3 out of the 12 possible signs are
compatible to your zodiac sign. If we assume that every person’s zodiac
sign can be any of the twelve signs, what is the probability that in some
random set of 4 people, you would be compatible with X=0, 1, 2, 3 or
all 4 of them?

N=4
p= P(compatible)= 3/12= 0.25 For x= 0,1,2,3,4
q= 0.75 the probabilities are
P(0) = 0.31640625
P(1) = 0.421875 P(3) = 0.046875
P(2) = 0.2109375 P(4) = 0.00390625

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
• Consider four multiple-choice questions of which you
have no knowledge. What is the probability of getting
exactly 3 questions correct if each question has 4
choices each that you guessed the answers to.

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
• If you are given 10 attempts to bet on a casino game that has a 1/5
chance of letting you win each game, what is the probability that you
will win at least twice?

P(X≥2) = 1- P(X≤1)

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
The Negative Binomial Distribution
The negative binomial RV and distribution are based on an experiment
satisfying the following conditions:
1. The experiment consists of a sequence of independent trials.
2. Each trial can result in either a success (S) or a failure (F).
3. The probability of success is constant from trial to trial
4. The experiment continues (trials are performed) until a total of K successes
have been observed, where K is a specified positive integer between 0 to X.

• Random variable X is the ordinal number of trial when the Kth success occurs.

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Negative Binomial Distribution
Parameters: K= number of successful attempts;
p = probability of “success” per attempt
q = (1-p) = P(failure) per attempt
Random Variable X = the ordinal number of Kth success among the X
attempts. The Kth success should occur on attempt no. X.

Probability Distribution Function :

Mean E(X) = k/p Variance =k(1-p)/p2


Siy, Discrete Probability Distributions, DLSU Industrial
Engineering Dept
Negative Binomial Example
• A pediatrician wishes to recruit 5 couples, each of whom is expecting
their first child, to participate in a new natural childbirth regimen. Let
p = P(a randomly selected couple agrees to participate).

• If p = .2, what is the probability that 15 couples must be asked before


5 are found who agree to participate? That is, with S = {agrees to
participate}, what is the probability that 10 F’s occur before the fifth
S?

Substituting k = 5, p = .2 , and x = 15

Ans: 0.314
Siy, Discrete Probability Distributions, DLSU Industrial
Engineering Dept
5th success on the 15th attempt

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
• Your basketball freethrow percentage is constant at 33.3%. If you
keep shooting freethrows, what is the probability that you take 7
shots before you make your 3rd made? (8th attempt is the 3rd success).

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Geometric Distribution
• Special case of the Negative binomial distribution when k=1, or first success in X
attempts.
Parameters: p = probability of “success” per attempt
q = (1-p) = P(failure) per attempt
Random Variable X = the ordinal number of first (K=1) success among the X attempts.
The first success should occur on attempt no. X.

Probability Distribution Function :

Mean E(X) = 1/p Variance =(1-p)/p2


Siy, Discrete Probability Distributions, DLSU Industrial
Engineering Dept
• Bains has had a string of bad luck on his search for a date. Suppose
that about 10 % of girls would be interested to date him, what is the
probability that the fifth girl she asks out would say yes?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
• Your basketball freethrow percentage is a decent 72%. What is the
probability that your third attempt is the first to go in?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Multinomial Distribution
•n = number of events
•n1 = number of outcomes, event 1
A multinomial experiment is almost identical
with the binomial with one main difference: a •n2 = number of outcomes, event 2
binomial experiment can have two outcomes, •nx = number of outcomes, event x
while a multinomial experiment can have •p1 = probability event 1 happens
multiple outcomes. •p2 = probability event 2 happens
•px = probability event x happens

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Multinomial Distribution
•n = number of events
•n1 = number of outcomes, event 1
Blood types have the following distribution: •n2 = number of outcomes, event 2
Type 0 = 45%, Type A= 35%, Type B= 15% and •nx = number of outcomes, event x
Type AB= 5% •p1 = probability event 1 happens
•p2 = probability event 2 happens
•px = probability event x happens
If 5 people are randomly selected, and their
blood types are ascertained, what is the
probability that there would be 2 blood type
O’s and one of each of the other types? 5!
( 0.45 )2 ( 0.35 )1 ( 0.15 )1 ( 0.05 )1
2! .1 ! 1 ! 1 !

=0.0319

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
The Hypergeometric Distribution
The assumptions leading to the hypergeometric distribution are as follows:
1. The population or set to be sampled consists of N individuals, objects,
or elements (a finite population).
2. Each individual can be characterized as a success (S) or a failure (F), and
there are k successes in the population.
3. A sample of n individuals is selected without replacement in such a way
that each subset of size n is equally likely to be chosen.

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
The Hypergeometric Distribution
• The random variable of interest is X = the number of k’s in the sample
of n objects.

• The probability distribution of X depends on the parameters n, M, and


N, so we wish to obtain P(X = x) = h(x; n, k, N).

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Hypergeometric Distribution
Parameters: k= number of successes in the population ;
n= number of sampled objects from the population,
sampled without replacement; (Binomial if sampled with replacemt)
N = size of the population, inclusive of the K kind

Random Variable X = the number of K typed objects among the


sampled n objects.
Probability Distribution Function :

Mean E(X) = Variance =

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Hypergeometric example
• You have 8 biogesic tablets mixed up among your medical bottle for
headaches that has a total of 50 tablets. Suppose that a random
selection of 3 tablets are made, what is the probability that there is
exactly 1 biogesic tablet in the three?

= =

=0.3514285714

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Hypergeometric Example
• During a particular period a university’s information technology office
received 20 service orders for problems with printers, of which 8 were
laser printers and 12 were inkjet models. A sample of 5 of these service
orders is to be selected for inclusion in a customer satisfaction survey.

• Suppose that the 5 are selected in a completely random fashion, so


that any particular subset of size 5 has the same chance of being
selected as does any other subset. What then is the probability that
exactly x (x = 0, 1, 2, 3, 4, or 5) of the selected service orders were for
inkjet printers?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
cont’d

• Here, the population size is N = 20, the sample size is n = 5, and the
number of S’s (inkjet = S) and F’s in the population are k = 12 and N –k
= 8, respectively.
Consider the value x = 2. Because all outcomes (each consisting of 5
particular orders) are equally likely,

P(X = 2) = h(x; n,k, N) = h(2; 5, 12, 20)

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
cont’d

• The number of possible outcomes in the experiment is the number of


ways of selecting 5 from the 20 objects without regard to order—that
is, . To count the number of outcomes having X = 2, note that there
are ways of selecting 2 of the inkjet orders, and for each such way
there are ways of selecting the 3 laser orders to fill out the sample.

• The probability numerator gives as the number of outcomes


with X = 2, so

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
The Hypergeometric Distribution
• In general, if the sample size n is smaller than the number of
successes in the population (k), then the largest possible X value is n.

• However, if k < n (e.g., a sample size of 25 and only 15 successes in


the population), then X can be at most k.

• Similarly, whenever the number of population failures (N – k) exceeds


the sample size, the smallest possible X value is 0 (since all sampled
individuals might then be failures).

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
The Hypergeometric Distribution
• However, if N – k < n, the smallest possible X value is
n – (N – k).

• Thus, the possible values of X satisfy the restriction


max (0, n – (N – k))  x  min (n, k).

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
• You draw seven cards from a well-shuffled deck of playing
cards, What is the probability that you have a trio of court
cards (King, Queen or Jack or any combination of three of
these)?

+ +++ =

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Multivariate Hypergeometric Distribution
If N items can be partitioned into k cells with a1, a2, ….. aK elements, the
the probability of the random variables X1, X2, … Xk representing the
number of elements selected from the subtypes a1, a2…..ak kinds of
objects, in a random sample of size n is given by:

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Multivariate Hypergeometric Distribution
• 5 cards are randomly chosen from a well-shuffled deck of playing
cards. What is the probability that the 5-card hand contain three
kings, an Ace and a Jack?

= (not likely)

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Poisson Distribution
• Distribution of “occurrence” events.
Random variable X = the number of occurrences given some expected density of (occurrences/time) or
(occurrences/area-space)

Examples of occurrences:
• Number of calls that randomly arrive at a calling center every minute when the average number of calls per minute
is known from historical data.
• Number of customers that arrive at a fastfood for a given hour (ex. A Tuesday 3pm to 4 pm), given some known the
number of customers that arrive every hour every Tuesday afternoon)
• Number of data packets sent to a cell site per day.
• Number of typographical errors in an issue of the La Sallian.
• Number of scratches found on a traded UpperDeck collectible card.
• Number of dandruff flakes found on the shoulder of the next person you see. Try also pimple, mole, bruisemarks,
blemishes,etc found on a person’s body. When how many should be expected for members of a population.

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Poisson Distribution
Input parameter: m = expected number of occurrences for a given interval of time or area-space
(density should be proportional to time or space)
= (density in occurrence/time) x (given time) or
(density in occurrence/area-space) x (given area-space)

 x e 
Probability distribution function: p ( x;  ) 
for X=0,1,2.. a
x!

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Lunchtime customers arrive at a fastfood counter following a Poisson distribution with an
expected 5 customers per minute. What is the probability that 2 customers will arrive in the next
30 seconds?

Let X= number of customers that arrive in the next 30 seconds.


= expected number of customers per 30 minutes.
 customers  1 min 
= proportionally  5  x30 sec  =
 min  60 sec 

= 2.5 customers in 30 secs

2.5 x e 2.5
Poisson pdf : p ( x; 2.5) 
x!
2.5 2 e 2.5
P( x  2)  p(2 ; 2.5)   0.2565
2!
Ans: There is a 25.65% probability that there would be exactly 2 customers in the next 30
seconds.

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
During the rainy season in the Philippines, it is expected that 3.2
tropical storms would occur during the month of August.
What is the probability that in a certain year’s August:
1. Exactly 4 storms form?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
During the rainy season in the Philippines, it is expected that 3.2
tropical storms would occur during the month of August.
What is the probability that in a certain year’s August:
1. Exactly 4 storms form?

2. From 2 to 5 storms form?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
During the rainy season in the Philippines, it is expected that 3.2
tropical storms would occur during the month of August.
What is the probability that in a certain year’s August:
1. Exactly 4 storms form?
2. From 2 to 5 storms form?

3. At most 3 storms occur?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
During the rainy season in the Philippines, it is expected that 3.2
tropical storms would occur during the month of August.
What is the probability that in a certain year’s August:
1. Exactly 4 storms form?
2. From 2 to 5 storms form?
3. At most 3 storms occur?

4. At least 4 storms occur?

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Poisson Approximation to the Binomial:
When n is sufficiently large, and the binomial proportion of success p is small, we can
use the Poisson p.d.function to approximate the binomial probability function. We let
Poisson parameter m be equal to the binomial expected value np (i.e. m=np).

x  np
(np) e
b ( x , n, p )    p q
n
x
x n x
 P( x; np) 
x!

if binomial pdf is computationally


difficult (e.g. n is large (>>100) and p is
small (e.g. p<0.05))

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Illustrative example:

The average number of people with AIDS in a certain country is about 2 every million. If
a survey was done such that of 3 million people truthfully reported their health status, What is
the probability that at most 4 people would report having AIDS?

Ans: Let X=number of people with AIDS in the sample of 3 million people. We have a
binomial probability function with n=3,000,000, p=2/106.

We are asked to get P(X4). We have b(x; 3M, 2/1M)

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept
Illustrative example:

The average number of people with AIDS in a certain country is about 2 every million. If
a survey was done such that of 3 million people truthfully reported their health status, What is
the probability that at most 4 people would report having AIDS?

Ans: Let X=number of people with AIDS in the sample of 3 million people. We have a
binomial probability function with n=3,000,000, p=2/106.

We are asked to get P(X4). We have b(x; 3M, 2/1M)

Siy, Discrete Probability Distributions, DLSU Industrial


Engineering Dept

You might also like