SPECIAL DISTRIBUTIONS

Probability distributions that are

commonly observed in reality.
 INTRODUCTION
• The behavior of a random variable can be
described by a probability density/mass
function, which gives the likelihood of an
occurrence of an outcome.
• Probability density/mass function of a
random variable will now be used
interchangeably with the term Probability
distribution.
 INTRODUCTION
• In real life situations, it has been observed
that several random variables behave similarly
in terms of their probability distribution.
– Graphically, the shape of the distribution has
similar features.
– Mathematically, the density/mass functions can
be summarized in one closed form
– Specific differences in the distribution is only due
to differently valued parameters.
 Example
• The Normal Distribution is a commonly observed
probability distribution differentiated by two
parameters (mean and variance).
 INTRODUCTION

• Saying that “X is normally

distributed” means that X is a
random variable that follows a
normal distribution or it has a
probability density function of a
normal distribution.
 Example
• In fact, all normally distributed random
variables have a probability density function as
follows:

1 𝑥−𝜇 2
−
𝑓𝑋 𝑥 = 𝑒 2𝜎 2
2𝜋𝜎 2
• Where 𝐸 𝑋 = 𝜇 and 𝑉𝑎𝑟 𝑋 = 𝜎 2
• 𝜇 and 𝜎 2 are the parameters of a normal
distribution
 INTRODUCTION
• Parameters are constants that define
the specific form of a probability
distribution.
• If we treat our probability distribution as
our statistical population, then
parameters of probability distributions
will have a similar notion as before.
 SPECIAL DISTRIBUTIONS
• These commonly observed
probability distributions shall be
termed as “Special Distributions”
• i.e. Poisson Distribution, Uniform
Distribution, Binomial Distribution
etc.
 FOCUS

•Normal Distribution
•Binomial Distribution
 NORMAL DISTRIBUTION

Bell-shaped Probability Distribution

 NORMAL DISTRIBUTION
• The Normal Distribution:
–It is a bell-shaped curve.
–It is symmetric about its mean.
–It has tails that extend on both ends
but it doesn’t touch the x-axis
–The area below the curve and above
the x-axis is 1.
NORMAL DISTRIBUTION
 NORMAL DISTRIBUTION

1 𝑥−𝜇 2
−
𝑓𝑋 𝑥 = 𝑒 2𝜎 2
2𝜋𝜎 2
• Where 𝐸 𝑋 = 𝜇 and 𝑉𝑎𝑟 𝑋 = 𝜎 2
2
• 𝜇 and 𝜎 are the parameters of a
normal distribution
 NORMAL DISTRIBUTION
 Standard Normal Distribution
 A normal distribution with mean 0 and
variance 1.
 A standard normal random variable
 Is a normally distributed random variable
with an expected value equal to 1 and
variance equal to 1.
 Commonly denoted as Z
 NORMAL DISTRIBUTION
• Theorem:

If X is a normally distributed random variable

with mean 𝜇 and variance 𝜎 2 and
𝑋−𝜇
z=
𝜎
Then Z is a standard normal random variable.
 Normal Distribution
If 𝑋~𝑁𝑜𝑟𝑚 𝜇, 𝜎 2 𝑜𝑟 𝑋~𝑁 𝜇, 𝜎 2 𝑎𝑛𝑑 "𝑎" 𝑖𝑠 𝑎 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡

𝑎−𝜇
𝑡𝑕𝑒𝑛 𝑃 𝑋 ≤ 𝑎 = 𝑃 𝑍 ≤
𝜎

𝑃 𝑋 ≤𝑎 =𝑃 𝑋−𝜇 ≤𝑎−𝜇
𝑋−𝜇 𝑎−𝜇
𝑃 𝑋≤𝑎 =𝑃 ≤
𝜎 𝜎
𝑎−𝜇
𝑃 𝑋≤𝑎 =𝑃 𝑍≤
𝜎
HOW TO READ A STANDARD NORMAL TABLE?
 APPLICATIONS
• Biology: most physically expressed traits of
living organisms are normally distributed. New
species evolve when the distributions of a
physical trait gradually splits into different
normally distributed traits, arising to
differently characterized populations.
• Which is of the same case in Demography in
the analysis of a changing population
APPLICATIONS
 APPLICATIONS
• Six Sigma is a set of tools/techniques in
manufacturing that aim near to perfection. The
chance of defect in a six sigma process must be less
than or equal to 0.00034%.
 COMPUTING PROBABILITIES USING THE
STANDARD NORMAL TABLE
PROBLEM:
IQ is described to be normally distributed. The
population average of IQ is 105 with a standard
deviation of 20. What is the probability of
randomly selecting a person in the population
with an IQ that is less than or equal to 80?
 COMPUTING PROBABILITIES USING
THE STANDARD NORMAL TABLE
GENERALLY, if 𝑋~𝑁 𝜇, 𝜎 2 then
𝑎−𝜇
𝑃 𝑋≤𝑎 =𝑃 𝑍≤
𝜎
 COMPUTING PROBABILITIES USING THE
STANDARD NORMAL TABLE
PROBLEM:
In a factory of light bulbs, the manufacturer
claims that their light bulbs has a continuous life
span of 850 days or better 99.90% of the time. If
the life span of a light bulb is observed to be
normally distributed with an average life span of
1000 days and a standard deviation of 50 days,
is the claim of the manufacturer true?
 COMPUTING PROBABILITIES USING
THE STANDARD NORMAL TABLE
GENERALLY, if 𝑋~𝑁 𝜇, 𝜎 2 then
𝑎−𝜇
𝑃 𝑋 ≥𝑎 =1−𝑃 𝑍 <
𝜎

= -
 COMPUTING PROBABILITIES USING THE
STANDARD NORMAL TABLE
PROBLEM:
The average height of a pine tree is 50 ft with a
standard deviation of 3 ft. The height of a pine
tree follows a normal distribution. What is the
probability that a randomly sampled tree has a
height between 47 ft and 67 ft?
 COMPUTING PROBABILITIES USING
THE STANDARD NORMAL TABLE
GENERALLY, if 𝑋~𝑁 𝜇, 𝜎 2 then
𝑏−𝜇 𝑎−𝜇
𝑃 𝑎≤𝑋≤𝑏 =𝑃 𝑍≤ −𝑃 𝑍 <
𝜎 𝜎

= -
 COMPUTING PROBABILITIES USING
THE STANDARD NORMAL TABLE
2
if 𝑋~𝑁 𝜇, 𝜎 then
𝑎−𝜇
1. 𝑃 𝑋 ≤ 𝑎 = 𝑃 𝑍 ≤ 𝜎
𝑎−𝜇
2. 𝑃 𝑋 ≥ 𝑎 = 1 − 𝑃 𝑍 < 𝜎
3. 𝑃 𝑎 ≤ 𝑋 ≤ 𝑏 =
𝑏−𝜇 𝑎−𝜇
𝑃 𝑍≤ −𝑃 𝑍 <
𝜎 𝜎
 Percentiles
𝑡𝑕𝑒 𝐾 𝑡ℎ 𝑝𝑒𝑟𝑐𝑒𝑛𝑡𝑖𝑙𝑒, 𝑃𝐾 , 𝑖𝑠 𝑛𝑜𝑤 𝑑𝑒𝑓𝑖𝑛𝑒𝑑 𝑎𝑠
𝑎𝑠 𝑎 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎 𝑐𝑜𝑛𝑡𝑖𝑛𝑢𝑜𝑢𝑠 𝑟𝑎𝑛𝑑𝑜𝑚
𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒, 𝑋, 𝑠𝑎𝑡𝑖𝑠𝑓𝑦𝑖𝑛𝑔:
Type equation here.

𝑃 𝑋 ≤ 𝑃𝐾 = 𝐾%

In the figure, b is equal

K% to the Kth percentile.
 Percentiles
• Notation:
𝑍𝛼 𝑖𝑠 𝑎 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑛𝑜𝑟𝑚𝑎𝑙
𝑟𝑎𝑛𝑑𝑜𝑚 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒 𝑠𝑢𝑐𝑕 𝑡𝑕𝑎𝑡:
𝑃 𝑍 ≥ 𝑍𝛼 = 𝛼

𝛼 Then, 𝑏 = 𝑍𝛼
 Percentiles
𝐻𝑒𝑛𝑐𝑒, 1 − 𝛼 100𝑡ℎ 𝑃𝑒𝑟𝑐𝑒𝑛𝑡𝑖𝑙𝑒, 𝑖𝑛 𝑎
𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑛𝑜𝑟𝑚𝑎𝑙 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛:
𝑃 1−𝛼 100 = 𝑍𝛼

𝛼 Then, 𝑏 = 𝑍𝛼
 Example
Using a personality test instrument, it
is found out that the emotional
personality index (score) among pre-
teens is normally distributed. On the
average a pre-teen will get a score of
10, while the variance is equal to 144.
What is the 67th percentile score?
 Example
• X is the emotional index
• 𝑋~𝑁 10,144 , 𝑤𝑕𝑎𝑡 𝑖𝑠 𝑡𝑕𝑒 67th percentile of X?
𝑁𝑒𝑒𝑑 𝑡𝑜 𝑓𝑖𝑛𝑑 𝑥 𝑠𝑢𝑐𝑕 𝑡𝑕𝑎𝑡 𝑃 𝑋 ≤ 𝑥 = 0.6700
𝑋 − 10 𝑥 − 10
𝑃 ≤ = 𝑃 𝑍 ≤ 𝑧 = 0.6700
12 12
𝑏𝑢𝑡, 𝑃 𝑍 ≤ 0.44 = 0.6700
𝑋 − 10
𝑃 ≤ 0.44 = 𝑃 𝑋 ≤ 𝟏𝟓. 𝟐𝟖 = 0.6700
12
 Example
For toddlers (around 3 yrs old), the
average weight is equal to 30 lbs with
a standard deviation of 6 lbs. The
weight is found to be normally
distributed. What is the 5th and the
95th percentile weights for toddlers?
 Example
• X is toddler’s weight
𝑋~𝑁 30,62 , 𝑤𝑕𝑎𝑡 𝑖𝑠 𝑡𝑕𝑒 5th and 95th
percentiles of X?
𝑃 𝑍 ≤ 1.645 = 0.95, 𝑃 𝑍 ≤ −1.645 = 0.05
𝑋 − 30
𝑃 ≤ 1.645 = 𝑃 𝑋 ≤ 39.87 = 0.95
6
𝑋 − 30
𝑃 ≤ −1.645 = 𝑃 𝑋 ≤ 20.13 = 0.05
6
BINOMIAL DISTRIBUTION

SUCCESS OR FAILURE
 Binomial Distribution
• It is common in studies that we ask questions
with two possible choices:
• What’s your stand? __ Pro __Anti
• Sex: __Male __Female
• Marital Status: __Married __ Not Married
• Are you voting for the candidate? __yes __no
• Are you satisfied with the performance of the
president? __yes __no
 Binomial Distribution
• Binomial distribution is one of the
common/special distributions that can fit a
random variable with two possible values
(dichotomous random variable).
• Binomial distribution applies if the random
variable is realized from a binomial random
experiment (Binomial Experiment)
 Binomial Experiment
• It consists of observing outcomes of a
sequence of n trials
• Each trial can only result to two possible
outcomes – “Success” or “Failure”
• The probability of “success”, p, is the same for
each trial in the sequence of n trials.
• Each trial is independent of the other. The
outcome of a trial does not affect the other.
 Examples of Binomial Experiment
• Tossing a coin
• Simple Random Sampling with replacement
(male or female, yes or no etc.)
• Simple Random Sampling without
replacement where the sample size is too
small compared to the population size (i.e. a
sample of 200 households from a population
of 1,000,000 households)
 Binomial Distribution
• If we have a binomial experiment and we
define a random variable X as:
𝑋 𝑡𝑕𝑒 𝑛𝑜. 𝑜𝑓 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑜𝑢𝑡𝑐𝑜𝑚𝑒𝑠 𝑖𝑛 𝑛 𝑡𝑟𝑖𝑎𝑙𝑠
Then, X is a binomial random variable
(binomially distributed).
• The binomial distribution has two parameters
(n “no. of trials”, p “probability of success”)
 Binomial Distribution
• 𝑖𝑓 𝑋~𝐵𝑖 𝑛, 𝑝 , 𝑡𝑕𝑒𝑛
𝑃 𝑋=𝑥
𝑛 𝑥 𝑛−𝑥
𝑝 1−𝑝 𝑖𝑓 𝑥 = 1,2,3, … 𝑛
= 𝑥
0 𝑜𝑡𝑕𝑒𝑟𝑤𝑖𝑠𝑒
Note:
𝑋 𝑖𝑠 𝑡𝑕𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑖𝑛 𝑛 𝑡𝑟𝑖𝑎𝑙𝑠
𝑛 𝑖𝑠 𝑡𝑕𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑟𝑖𝑎𝑙𝑠
𝑝 𝑖𝑠 𝑡𝑕𝑒 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑠𝑢𝑐𝑐𝑒𝑠𝑠
 Example
• If the probability of getting a head in tossing a
coin is 0.75, what is the probability that heads
will appear 3 out of 5 tosses?

5 3 5−3
𝑃 𝑋=3 = 0.75 0.25 = 0.2637
3
 Example
• An Ice-cream brand claims that 80% of their
many customers will purchase their new
flavor: chili-rambutan ice-cream. They took a
random sample of 15 customers and asked
them if they will purchase a chili-rambutan
ice-cream. We define a random variable as the
number of customers in the sample, who said
they will purchase. Compute the probabilities
for the different possible values of the random
variable. If only 3 out of the 15 said that they
will purchase, should you believe the claim of
the ice-cream brand?
 Binomial Distribution
• 𝑖𝑓 𝑋~𝐵𝑖 𝑛, 𝑝 ,
𝑡𝑕𝑒𝑛 𝐸 𝑋 = 𝑛𝑝 𝑎𝑛𝑑 𝑉𝑎𝑟 𝑋 = 𝑛𝑝𝑞
• 𝑞 =1−𝑝
 Binomial Distribution
• If only 10% of the population knows the tv
show called “The Show”, how many are
expected to know “The Show” in a random
sample of 30 persons?

• Answer: 3