Topic 9: Discrete Probability Distribution: Binomial Probability

Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Describe the characteristics of the binomial probability distribution (experiment)

 Compute probabilities using Binomial Probability distribution

Topic outline

1. Properties or assumptions of binomial probability distribution (experiment)

2. Computing binomial probabilities
3. Synopsis
4. Wrap up discussion questions
5. Next session’s assignment

Reading Assignment Discussion:

 List the properties (assumptions) of Binomial probability distribution/

experiment
 How are probabilities computed in Binomial experiment?

Reading Text:

Jakob Bernouli, Swiss mathematician, published article on the binomial theorem. The
binomial probability distribution is one of the most widely used discrete probability
distributions. It has many applications. It is used to find the probability that an outcome
will occur x times in n experiments (trials).

Properties or assumptions of a binomial experiment or probability distribution:

 It is discrete probability distribution

 Trials are identical, independent and random.
  There are fixed number off trials (n trials or sample size).
 There are only two mutually exclusive outcomes of the trial designated as
success or failure.
 The probability of success ‘P’ is uniform throughout the n trials
 It is assumed that sampling is with replacement from large population (n/N≤5%)

Example 7 of the topic for Session 7 is a typical example of binomial distribution or

experiment. Number of heads is a discrete random variable. 3 identical trials are run;
the outcome of one does not affect the outcome of the other, and results are unknown
but possibilities are known, thus trials are independent and random. In each toss there
are 2 mutually exclusive outcomes Head the success and Tail the failure event. The
probability of Head (success) ‘P’ is 0.5), and the probability of tail (failure) ‘Q=1-P’ is
0.5 in each trial (toss) respectively; and P & Q remain the same for all trials. A binomial
probability is computed by the formula:

; Q=1-P

Note:
 x is the random variable defined as the number of successes
 n-x is the number of failures
 C denotes combination
 n is the number of trials or sample size
 P is the probability of success on each trial; 1-P or Q is the probability of failure

n & P are the two binomial parameters that are required to find probability of x
successes in n trials (random sample size) for a binomial experiment. Probability Tables
can also be used for limited number of n and P to find binomial probabilities.

Example 9

1. Construct the binomial probability distribution for number of heads for the
binomial experiment of tossing 3 coins.
 Given: Success=number of heads, n=3, P=0.5, Q=0.5;
The probability of exactly 0 (no) head in tossing 3 coins, P(x=0)=?
= = 1*1*0.125=0.125.
Likewise, =3*0.5*0.25=0.375
= 3*0.25*1=0.375
= 1*0.125*1=0.125
Note: For example 3C2=3, gives the frequency or possibilities or number of ways or
arrangements of getting 2 heads in 3 tosses i.e. THH, HTH, HHT.
9.1 The probability distribution for the binomial experiment of tossing 3
coins for number of heads
Xi (# of Heads) 0 1 2 3 Total
P(Xi) 0.125 0.375 0.375 0.125 ∑P(Xi)=1

2. The probability that a randomly chosen sales prospect will make a purchase is
0.20. If a sales representative calls on fifteen prospects, find the probability that:
a. Fewer than two sales are made
Given: n=15, P=0.2, Q=0.8, P(x<2)=?
P(x<2)=P(x=0)+P(x=1)=[15C0*0.20*0.815-0+15C1*0.21*0.815-1]=
=0.0352+0.1319= 0.1671
b. At least two sales are made
P(x≥2)=P(x=2,3,…15)=1-P(x<2)=1-[P(x=0)+P(x=1)]=1-0.1671=0.8329
(refer to a)

Exercise 9
1. A manufacturer of window frames knows from long experience that 5% of the
production will have some type of minor defect that will require an adjustment.
What is the probability that in a sample of 20 window frames:
a. None will need adjustment?
b. At least one will need adjustment?
c. More than two will need adjustment?
 2. In a recent study, 90% of the homes in Bole area were found to have large-
screen TVs. In a sample of nine homes, what is the probability that:
a. All (exactly) nine have large-screen TVs?
b. Less than five have large-screen TVs?
c. More than five have large-screen TVs?
d. Develop a binomial probability distribution for the problem

Synopsis
 A binomial probability distribution or experiment:
o It has n trials that are similar, independent and random
o Probability of success (P) remain the same for series of trials
o Random sampling is with replacement form large population
o Binomial parameters are n & P; then P(x)= probability of x successes in n
trials or sample size:

o ; Q=1-P

Wrap up Discussion Questions:

 Briefly explain a binomial experiment and a binomial random variable.

 What are the parameters of the binomial probability distribution? What do they
mean?
 List the attributes of a binomial experiment or distribution
 How do we calculate binomial probabilities?
 Explain the following binomial probability distribution properties: sampling with
replacement from large population, independence, and P remaining constant in
series of trials.

Next Session’s Assignment:

 Attempt exercise 9, #1 and 2
 Read about computing the mean, variance and standard deviation for a binomial
probability distribution.
Topic 10: Discrete Probability Distribution: The Expected Value (Mean)
and Standard Deviation for the Binomial Probability
Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Compute the expected value (mean), variance and standard deviation for the
binomial probability distribution

Topic outline

1. Computing expected mean for a binomial probability distribution

2. Computing variance and standard deviations of binomial probability distribution
3. Synopsis
4. Wrap up discussion questions
5. Next session’s assignment

Reading Assignment Discussion:

 How do we compute the mean, variance, and standard deviation for a

binomial probability distribution?

Reading Text:

The mean (μ), the variance (σ2), and the standard deviation (σ) of a binomial
distribution are computed in a “shortcut” fashion as follows respectively:

; ;

Example 10

1. Find the mean, variance and standard deviation for the binomial probability
distribution in Example 7 and 8 of Session 7 and 8: n=3, P=0.5, Q=0.5
 E(X)=µ=3*0.5=1.5 heads; σ2=3*0.5*0.5=0.75 heads; σ= =0.87 heads
2. The probability that a randomly chosen sales prospect will make a purchase is
0.20. If a sales representative calls on fifteen prospects, find the expected
number of sales (as a long-run average), the variance, and the standard
deviation associated with making calls on 15 prospects.
= 15*0.2=3 sales

= 15*0.2*0.8=2.4 sales

= =1.55 sales

Exercise 10

1. A university has found that 20% of its students withdraw without completing the
first semester course. Assume that 6 students have registered for the course this
semester.
a. What is the probability that two or fewer will withdraw?
b. What is the probability that exactly four will withdraw?
c. What is the expected number of withdrawals?
d. What is the variance for the number of withdrawals?
e. What is the standard deviation for the number of withdrawals
2. If 5% of new automobiles require warranty service within the first year and X
sells 8 automobiles in a month
a. What is the probability that none of these automobiles requires warranty
service?
b. Find the probability that exactly two automobiles require warranty service?
c. Compute the mean and standard deviation of this probability distribution.

Synopsis

 For binomial probability distribution:

o ; ;
Wrap up Discussion Questions:

 How do we compute the mean and variance of a binomial probability

distribution?

Next Session’s Assignment:

 Attempt exercise 10, #1 and 2

 Read about the properties of hypergeometric probability distribution.

Topic 11: Discrete Probability Distribution: Hypergeometric Probability

Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Describe the characteristics of the hypergeometric probability distribution

 Compute probabilities using formula
 Find the mean, variance and standard deviation of a hypergeometric
distribution

Topic outline

1. Properties (assumptions) of hypergeometric probability distribution

2. Computing probabilities for hypergeomeric distribution
3. Mean and variance of hypergeometric probability distribution
4. Synopsis
5. Wrap up Discussion Questions
6. Next session’s assignment

Reading Assignment Discussion:

 List the characteristics of a hypergeometric probability distribution

 How do we calculate hypergeometric probabilities?
  How do we compute the mean, variance, and standard deviation?

Reading Text:

When sampling is done without replacement of each sampled item taken from a finite
population of items, the binomial process does not apply because there is a systematic
change in the probability of success as items are removed from the population.
Hence, the hypergeometric probability distribution is instead used.

Properties (assumptions) of hypergeometric probability distribution (experiment)

 It is a discrete probability distribution
 An outcome on each trial of an experiment is classified into one of two
mutually exclusive categories—a success or a failure.
 The random variable is the number of successes in a fixed number of trials.
 The trials are not independent.
 We assume that we sample from a finite population (n/N > 0.05 (5%)
without replacement
 So, the probability of a success changes for each trial.

For example, in choosing a random sample of 2 men for committee, from population of
10 men, without replacement; each man has a chance of 1/10 to be selected the first
round, & has 1/9 chance to be selected the second time, note that trials are dependent.
The formula for the hypergeomeric distribution is:

P(x)= SCx*(N-S)C(n-x)
NCn

 N= size of population
 n= size of sample
 S= # of successes in N or population;
 N-S= # of failure in N or population;
 X=# of successes in sample size of n;
 n-X= # of failure in n sample size
Note: Thus choosing successes (X from S is SCx ); choosing failures (n-X from N-S is N-

SCn-x), in choosing both at the same time we multiply both terms; and totally out of N

population size, n sample size can be chosen randomly and without replacement in NCn

ways.

Mean and Variance of hypergeometric Probability Distribution

 s   s  s  N  n 
Mean:µ= E ( X )  n  and Variance  x  n 1   
2

N  N  N  N  1 

Note: as the population is finite, the variance is multiplied by the finite population

 N n
correction factor  .
 N 1 
Example 11

1. Mr. X has received a shipment of 50 TV sets. 10 of these sets are found to be

defective. The owner decided to test a sample of 5 TV sets randomly.

Note that sampling is done without replacement and the sample size of 5 or 10% of the
population. This is greater than the 5% requirement. Given: N=50; n= 5; S= 10;

a. What is the probability that 1 of the 5 TV sets will be defective?

P(x)= SCx*(N-S)C(n-x) =P(x=1)= [10C1* (50-10)C(5-1)] / 50C5 =

NCn
=913,900/2,118,760=0.431
b. What about 2 of them are defective?
P(x=2)= [10C2* (50-10)C(5-2)] / 50C5 = 444,600/ 2, 118,760=0.21
c. Develop the probability distribution for the indicated problem. (X represents
number of defective TV sets)
Table 11.1 Probability distribution for the number of defective TV sets in a
random sample of 5 TV sets taken from shipment of 50 TV sets
Xi 0 1 2 3 4 5 Total
∑P(Xi)≈1
0.3106 0.4313 0.2098 0.0442 0.004 0.0001
P(Xi)

2. If 180 of the 200 shipping orders that the inspector examined have authorized
credit approval. What is the mean and the variance of the number in a sample of
40 randomly chosen orders will have credit approval?
Given: n=40, N=200, S=180

 s 
Mean:µ= E ( X )  n  = Mean=E(X) = 40(180/200) = 36; and
N
 s  s  N  n 
Variance  x  n 1   
2

 N  N  N  1 
Variance=4(180/200)(20/200) (160/199) = 2.8945

Exercise 11

1. Suppose Mr. x has 20 delivery trucks. If 5 of the 20 trucks have suspension

problems and 3 trucks were selected at random to be tested,
a. What is the probability that one of those tested have suspension problem?
b. Find the mean and variance of number of trucks having suspension
problems in the random sample.
c. Construct the probability distribution for the problem.
2. A new flavor of toothpaste has been developed. It was tasted by a group of 10
people. 6 of the group said they liked the new flavor and the remaining 4
indicated they did not. 4 of the 10 are selected to participate in an in depth
interview. What is the probability that of those selected for the in depth interview
2 liked the new flavor and 2 did not?
3. A shipment of pins contains 25 good (non-defective) ones and 2 defective ones.
At the receiving department, an inspector picks three pins at random and tests
 them. If one defective pin is found among the three that are tested, the
shipment would be rejected.
 What is the probability that the shipment would be rejected?

Synopsis
 The hypergeometric probability distribution is useful for determining the
probability of a number of occurrences when sampling without replacement from
finite population (n/N>5%).
 It counts the number of successes (x) in n selections, without replacement, from
a population of N elements, S of which are successes and (N-S) of which are
failures

 P(x)= SCx*(N-S)C(n-x)
NCn
 s   s  s  N  n 
Mean=µ= E ( X )  n  and Variance  x  n 1   
2

N  N  N  N  1 

Wrap up Discussion Questions:

 What are the conditions that must be satisfied to apply the hypergeometric
probability distribution?
 Compare and contrast the hypergeometric and binomial probability distribution.
 How do we find probabilities in hypergeometric experiment?
 How do we compute the mean and variance of hypergeometric distribution?
 What must be n/N, if the population has to be small/finite enough, or sample is
large enough?

Next Session’s Assignment:

 Attempt exercise 11, #1-3

 Read about the properties of Poisson probability distribution.
Topic 12: Discrete Probability Distribution: Poisson Probability
Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Describe the properties of Poisson probability distribution,

 Compute Poisson probabilities using formula and table

Topic outline

1. Properties (Assumptions) of Poisson probability distribution

2. Computing Poisson probabilities
3. Synopsis
4. Wrap up discussion questions
5. Next session’s assignment

Reading Assignment Discussion:

 List the characteristics of a Poisson probability distribution
 What is the Poisson parameter?
 How do we calculate Poisson probabilities?
Reading Text:

This distribution is named after French mathematician Simeon Denise Poisson (1781-
1840). It is a discrete distribution. It expresses the probability of a number of
events/occurrences (successes) occurring in a fixed interval of time, space (area,
volume, distance) etc. For example, the random variable of interest might be the
number of arrivals at a foreign exchange service desk at a bank in one hour, the
number of accidents in 100 km of a highway, or the number of Walia Ibexes observed
in 2 hectares of the Semen National Park.
Properties (Assumptions) of Poisson Probability Distribution (Experiment)
 The probability of an occurrence (success) is the same for any two intervals
of equal length and remains proportional.
 The number of occurrences (successes) in any interval is random, and
independent of the number of occurrences (successes) in any other interval.
 The Poisson parameter is the mean number of occurrences (successes) i.e.
denoted by µ or λ remains constant (for similar intervals)

P(x)=The probability of x number of occurrences or successes in an interval is

computed as follows:

 Mu (µ) or lambda (λ) = expected value or mean (average or rate) number

of successes in an interval
 e = Approximately 2.7183

Poisson Probabilities for any x can easily be obtained from tables given the µ or λ
(Note: table is available only for limited mean)

Example 12

Customers arrive at a photocopying machine at an average rate of 2 every minute. The

number of arrivals is distributed according to a Poisson distribution. Find the following
probabilities:

a. One arrival in any minute

= =2*0.1353=0.2707

b. There will be three arrivals during any five minutes period (interval)
As the interval of the mean is changed from 1 minute to five minutes, adjust the
mean; it will be µ=10

P(x=3)= =(1000* 0.000045)/ 6= 0.0076

 c. Suppose the mean is 6 every minute, find the probability of more than 2 arrivals
in any minute (Use table) µ=6/1min.
P(x>2)=P(x=3)+P(x=4)…., it extends infinitely in Poisson case
P(x>2)=1-P(x≤2) or 1-P(x<3)=1-[P(x=0)+P(x=1)+P(x=2)]=1-
[0.0025+0.0149+0.0446]=1-0.062=0.938
d. Construct Poisson probability distribution for (0 through 10) number of arrivals in
any minute interval. Note: X represents number of arrivals in any 1 minute
interval, ( you can use table or formula to find probabilities)
Xi 0 1 2 3 4 5 6 7 8 9 10
P(Xi) 0.1353 0.2707 0.2707 0.1804 0.0902 0.0361 0.0120 0.0034 0.0009 0.0002 0.0000

Exercise 12

1. If the mean number of errors per day is 0.8,

a. What is the probability that there are less than 2 errors in a particular
day?
b. Construct probability distribution for the problem for number of errors
ranging from 0 through 8.
2. A telephone company’s goal is not have more than three line failures in a
particular 1km line. Currently the company is experiencing four line failures in
1km line on average.
a. What is the probability that the company will meet its goal?
b. What’s the probability that the company will not meet its goal?
3. During rush hours, accidents occur at the rate of 2 per hour. The morning rush
hour lasts for 1:30 minutes and the afternoon rush period lasts for 2:00 hours.
On a particular day, what is the probability that
a. There will be no accident during the morning rush hour?
b. Two accidents will occur during the afternoon rush hour?
c. Four or more accidents will occur during the morning rush hour?
d. No accident will occur during both morning and afternoon rush hour?
Synopsis

 A Poisson experiment (probability distribution) is characterized by the following

properties:
o The number of successes that occur in any interval is random, and
independent of the number of successes that occur in any other interval.
o The probability of a success in an interval is the same for all equal-size
intervals and changes proportionally as interval changes
o Only one parameter the mean (µ) is needed to find probabilities either
form formula or table
 P(x)=The probability of x number of occurrences or successes in an interval is

computed as follows:

Wrap up Discussion Questions:

 What are the conditions that must be satisfied to apply the Poisson probability
distribution?
 Identify the Poisson parameter
 How do we find Poisson probabilities?
Next Session’s Assignment:
 Attempt exercise 12, #1, 2, and 3
 Read about the Poisson approximation of a binomial probability distribution.
Topic 13: Discrete Probability Distribution: Poisson Approximation
of the Binomial Probability Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Compute binomial probabilities using Poisson distribution when n is large & p is

small;
 Find standard deviation of a Poisson distribution

Topic outline

1. Solving binomial probabilities using Poisson probability distribution

2. Mean, Variance, & Standard deviation of Poisson Probability Distribution
3. Synopsis
4. Wrap up discussion questions
5. Next session’s assignment

Reading Assignment Discussion:

 What are the conditions that must be satisfied to solve binomial probability
problems using Poisson probability distribution?
 How are variance and standard deviation calculated for Poisson distribution?

Reading Text:

When the number of observations or trials n in a binomial experiment is large,

computations are quite tedious. Further, tabled probabilities for very small values of P
are not generally available. Fortunately, the Poisson probability distribution can be used
as an approximation of the binomial probability distribution when the probability of
success (P) is small and the number of trials (n) is large. Simply find µ or  =nP and use
the Poisson tables. As a rule, the approximation will be good whenever P < 0.05 (5%)
and n > 20.
Example 13

A hotel has 50 rooms if 2 % of the hotel rooms are not occupied on a given day what is
the probability that less than 2 rooms will not be occupied?: (Apply Poisson
approximation of the Binomial distribution and use table)

Given: n=50 P=0.02 (2%); P(x<2)=?

As n≥20 and P≤5% is satisfied the binomial probability problem can be solved using
Poisson probability table by finding the mean (µ); µ=n*P=50*0.02=1.

Thus µ=1, P(x<2)=P(x=0)+P(x=1)= 0.3679+0.3679=0.7358

Mean, Variance, & Standard deviation of Poisson Probability Distribution

For the Poisson probability distribution, the mean and variance are equal to µ and the
standard deviation is equal to the square root of µ. i.e.

σ2 = µ

σ=
µ

Exercise 13

Note: Find the following binomial probabilities using Poisson (if possible)

1. For a large shipment of transistors from a supplier, 1 percent of the items is

known to be defective. If a sample of 30 transistors is randomly selected, what is
the probability that two or more transistors will be defective?
2. Suppose 2% of the products produced by MAD are defective. For a random
sample of 200 products, find the probability that (a) None of the products are
defective? (b) Three or more of the products are defective?
3. It is estimated that 0.5% of the passengers use telephone service. What is the
probability that of 1200 passengers, at least 5 use telephone service?
Synopsis
 Whenever P < 0.05 (5%) and n > 20; we can solve binomial probability
problems using Poisson probability distribution, find µ=nP and read
probability from table.
 For Poisson probability distribution: µ=σ2; σ=

Wrap up Discussion Questions:

 List the conditions that must be satisfied to approximate binomial probability

distribution using Poisson probability distribution.
 How do we compute binomial probabilities using Poisson?
 How is the variance and standard deviation computed for Poisson distribution?

Next Session’s Assignment:

 Attempt exercise 13 #1-3

 Read about the basics of a continuous probability distribution and its typology.

Topic 14: Continuous Probability Distribution: Introduction

Topic Learning Objectives:

By the end of this session students are expected to:

 Describe continuous random variable and continuous probability distribution;

 Indicate the different types of continuous probability distribution

Topic outline

1. Continuous random variables and continuous probability distributions

2. Synopsis
3. Wrap up discussion questions
4. Next session’s assignment
Reading Assignment Discussion:

 What is a continuous probability distribution?

 List some types of continuous probability distribution

Reading Text:

Unlike discrete random variables, a continuous random variable assumes any value in
one or more intervals. The possible values that a continuous random variable can
assume are infinite and uncountable. Example: the variable representing the time taken
by the worker to finish the work, the time taken to complete an examination, amount of
milk in a pot, prices of houses, the flight time of an air plane from Gondar to Addis
Ababa, the weight of a person, the life of a battery, etc. are examples of continuous
random variables. Because the number of values contained in any interval is infinite,
the possible number of values that a continuous random variable can assume is also
infinite. Moreover we cannot count these values.

Because there are an infinite number of possible fractional measurements, one cannot
list every possible value with a corresponding probability. Instead, a probability density
function is defined. This mathematical expression gives the function of x, represented
by the symbol f(x), for any designated value of the random variable x. The plot for such
a function is called a probability curve, and the area between any two points under the
curve indicates the probability of a value between these two points occurring by
chance.

Several standard continuous probability distributions are applicable as models to a wide

variety of continuous variables under designated circumstances. Probability tables have
been prepared for these standard distributions, making it unnecessary to use the
method of integration in order to determine areas under the probability curve for these
distributions. Some types of continuous probability distribution are namely: the Uniform
Probability Distribution, the Exponential Probability Distribution, the Normal Probability
Distribution, Student’s t Distribution, F Distribution, Chi-Square Distribution and etc. As
the normal probability distribution is the most widely known and used of all the
distributions, it is discussed extensively in subsequent sections; and the other
continuous probability distributions listed next to it above will also be discussed in their
respective sections.

Synopsis

 A continuous random variable assumes any value in one or more intervals. The
possible values that a continuous random variable can assume are infinite and
uncountable.
 The area between any two points under the probability curve indicates the
probability of a value between these two points occurring by chance.
 Probability tables of several standard continuous probability distributions are
used to find the probability of the range of values of a continuous random
variable
 There are many types of continuous probability distribution that are identified by
different names and properties
 Normal probability distribution is the most widely used of all

Wrap up Discussion Questions:

 Describe a continuous random variable

 What is a probability density function and probability curve?
 How is probability found for range of values of a continuous random variable
from a probability curve?
 List the different types of the continuous probability distribution
 Which one is the most widely used?

Next Session’s Assignment:

 Read about the properties of the normal probability distribution

Topic 15: Continuous Probability Distribution: Properties of the Normal
Probability Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 List and describe the characteristics of a normal probability distribution

Topic outline

1. The characteristics of the normal probability distribution

2. Synopsis
3. Wrap up discussion questions
4. Next session’s assignment

Reading Assignment Discussion:

 Briefly explain the features of a normal probability distribution

Reading Text:

The normal probability distribution is the most commonly used of all probability
distributions; it is important in statistical inference for three distinct reasons:
1. The measurements obtained in many random processes (naturally and in
industry) are known to follow this distribution. Some examples are the IQs,
weights, and heights of a large number of people and the variations in
dimensions of a large number of parts produced by a machine.
2. Normal probabilities can often be used to approximate other probability
distributions, such as the binomial and Poisson distributions.
3. Distributions of such statistics as the sample mean and sample proportion are
normally distributed when the sample size is large, regardless of the distribution
of the parent population
The characteristics of the normal probability distribution

 It is a continuous probability distribution

 It is Bell shaped
 Its mean, median, and mode are at the center and are equal
 It is Symmetrical about the mean (the left and right halves are mirror images
of each other; area is 0.5 on both sides of the mean) and is mesokurtic
(moderately peaked)
 It is unimodal (it has a single peak at the center)
 It is asymptotic (theoretically, the left and right tails of the curve extend
infinitely; it gets closer to the x axis but it does not touch it)
 The total area under the normal curve equals to 1 or 100%, the area under
the probability distribution curve of a continuous random variable between
any two points is between 0 and 1 inclusive. That is, the probability of x lies
in the range 0-1 inclusive.
o The probability of a continuous random variable x is given by the area
under the curve between the two limits of the interval. Thus, for the
normal probability distribution, the probability is always calculated for
an interval; P(a < x < b) = area under the curve from a to b
o The probability that a continuous random variable x assumes a single
value is always zero. This is because the area of a line which
represents a single point is zero. Since the probability of any single
value is zero, P (a < x < b) = P (a < x < b). The interval a < x < b
does not include a and b but the interval a < x < b includes a and b.
However, both of these intervals contain the same probability or area
under the normal curve.
o The proportion of the observation of a normal distribution or area
under the normal curve i.e. within: µ±1*σ is 68.26%, µ±2*σ is
95.44% & µ±3*σ is 99.72% (almost all)
  There are infinite number of normal distributions that are uniquely defined by
the two parameters the mean (µ) and standard deviation (σ). The location of
a normal distribution is determined by the mean, μ. The dispersion or spread
of the distribution (width of the curve) is determined by the standard
deviation, σ.
 The height of the density function, or probability curve, for a normally
distributed variable is given by:
1 x 2
1 (  2 )( ]
f (X )  e 
 2
Where, f(x) = height of the normal distribution curve
µ= Mean σ= Standard deviation
e= 2.71828  = 3.14159
However, we will not use this equation to find the area under a normal
distribution curve or probability. Instead we use the values presented in
the standard normal distribution (table).

-∞ +∞

Figure 15.1 Characteristics of a Normal Distribution

Synopsis

 The normal distribution is the most commonly used of all probability distributions
in statistical analysis.
  It is bell shaped, symmetrical and moderately peaked; and asymptotic
 There is a family of unlimited normal distribution, each defined by two
parameters: its mean (µ) and standard deviation (σ)

Wrap up Discussion Questions:

 What are the parameters that define normal distribution? How do they uniquely
define a normal distribution?
 List and describe the major features of a normal probability distribution
 Why is the normal distribution the most widely used of all probability distribution
in statistical analysis?

Next Session’s Assignment:

 Read about standard normal (z) distribution; and how other normal distribution
is converted to this form.

Topic 16: Continuous Probability Distribution: The Standard Normal

(Z) Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Explain the characteristics of the standard normal (z) distribution;

 Find area under the standard normal distribution curve from the table;
 Convert a normal distribution to the standard normal distribution

Topic outline

1. Finding normal probabilities using the standard normal distribution /Z distribution

(table)
2. Synopsis
3. Wrap up discussion questions
 4. Next session’s assignment

Reading Assignment Discussion:

 Briefly explain the characteristics of the standard distribution and its use

Reading Text:

Since every different combination of µ and σ would generate a different normal

probability distribution (all symmetrical and mesokurtic), tables of normal probabilities
are based on one particular distribution: the standard normal distribution. This is the
normal probability distribution with mean (µ= 0) and standard deviation (σ=1). 99.72%
of the standard normal curve falls between points that are within ±3*σ from the mean
(µ) i.e. (between -3 and +3); the total area under the curve is 100% or 1.

Any value x from a normally distributed population can be converted into the equivalent

standard normal value z by the formula: .

A z value restates the original value x in terms of the number of units of the standard
deviation by which the original value differs from the mean of the distribution. A
negative value of z would indicate that the original value x was below the value of the
mean, and a positive value of z would indicate the original value x was above its mean.
The standard normal (z) distribution (table) indicates proportions of area for various
intervals of values for the standard normal probability distribution, with the lower
boundary of the interval always being at the mean. Converting designated values of the
variable x into standard normal values allows finding normal probabilities using the z
table easily, and it makes it unnecessary to use the complex method of integration with
respect to the equation for the density function shown in previous section.
Using Tables:

Divide the number and decimals of the z values as follows: The first digit before the
decimal point and the first decimal point are taken and checked for at the left most
column, and the second and third components of the decimal digits at the top most row
and the intersection in the body of the table will be the corresponding area (probability)
for the +z value measured starting from 0.

Example 16

1. Find the area (probability) under the standard normal curve by using the z table:
Note: For each of the questions below, locate the area under the following curve
(Fig. 16.1) for each of the following ranges of z values

µ=
Figure 16.1 The Standard Normal Distribution

a. Find the area between 0 and +z or 0 and -z directly from table:

E.g. Area between z=0 & z=+1.95 or z=-1.95, divide 1.95 into 1.9 and
.05, and read 1.9 on the left most column, and .05 on the top most
row; the decimal at the intersection of 1.9 and .05 in the body of the z
table is the area (probability) between 0 and 1.95
P (0<z<1.95) or P(-1.95<Z<0)= 0.4744
b. To find area to right of +z or to the left of -z, find the area between 0 and ± z
and subtract it from 0.5;
E.g. P (z >1.95) or P(z<-1.95)=0.5-0.4744=.0255
 c. For area between two points on different sides of the mean: add these two
areas
E.g. P (-1.95<z<2.12)=0.4744+0.4830=0.9574;
d. For area above –z or below +z continuing indefinitely, identify area
(probability) between 0 and z and then add to it 0 .5;
E.g. P(z>-1.95),or P(z<1.95)= 0.5+0.4744=0.9744
e. When 2 points are on the same side of the mean, first find the areas
(probability) between the mean and each of the 2 points, and then subtract
the smaller from the larger area.
E.g. P(1.95<z<2.12)=0.4830-0.4744=.0086
f. If z value is greater than 3.99 probability (area) between 0 and ±z is
approximated to 0.5.
E.g. P (0<z<4)=almost 0.5; P(z>-20)= almost 1; P(z<-6)=almost 0
2. Conversion of values of normal random variables in x scale to values of the
standard normal distribution z score(scale) the following conversion formula is

used:

E.g. If a normally distribution of grades has a mean of 50 and standard deviation

of 5, what is the probability that a grade lies between 40 and 50, or what
proportion of the grades lie between 40 and 50?
P (40<x<50)=P(-2<z<0),
N.B.: x is converted to z; z=(40-50)/5=-2, & z=(50-50)/5=0
To find the area (probability) between -2 and 0; we find the equivalent
area between 0 and +2, and i.e. 0.4772

Exercise 16

1. Find the area under the standard normal curve

a. between z = 0 to z = 2.34 d. between z = -2.15 and z = 2.37
b. between z = -2.15 and z = -2.37 e. to the right of z= 1.83
c. to the left of z = 0.79 f. to the right of z =5.43
 2. Determine the following probabilities for the standard normal distribution.
a. P(1.82 < z < 2.35) d. P(-2.41 < z < 1.41)
b. P(z > 4.45) e. P(z >-1.26)
c. P(z<11) f. P(-8<z<0)
3. For a standard normal distribution find the area within 1.5 standard deviation
of the mean

Synopsis

 The standard normal (z) probability distribution is a type of normal distribution

that has a mean of 0 and a standard deviation of 1.
 Any normal probability distribution can be converted to the standard normal

probability distribution by the following formula:

 By standardizing a normal probability distribution, we can report the distance of

a value from the mean in units of the standard deviation.

Wrap up Discussion Questions:

 Explain the use and features of the standard normal (z) distribution

Next Session’s Assignment:

 Attempt exercise 16, #1-3

 Read about computing normal probabilities using the z distribution
Topic 17: Continuous Probability Distribution: Computing Normal
Probabilities

Topic Learning Objectives:

By the end of this session students are expected to:

 Determine probabilities for range of values (observations) of normal random

variable using the standard normal (z) distribution

Topic outline

1. Computing normal probabilities using the Z distribution

2. Synopsis
3. Wrap up discussion questions
4. Next session’s assignment

Reading Assignment Discussion:

 Explain how normal probabilities are calculated using the z distribution for
any interval of values of normal random variable

Reading Text:

The standard normal distribution is very useful for determining probabilities for any
normally distributed random variable. The basic procedure is to find the z value for a
particular value of the random variable based on the mean and standard deviation of its
distribution.
The z value can be found using the conversion formula:

. Then, using the z value, we can use the standard normal distribution to find

various probabilities. The following example/solution describes the details of the

application.
Example 17
The IQ score of students is normally distributed with mean of 120 and standard
deviation of 20. Find the following probabilities:

a. What proportion of students has IQ between 100 & 130?

P (100<x<130)=?

Convert x to z; = and

P (-1.00<z<0.50) = 0.3413+0.1915=0.5328

Note: From the z table, the area under question in the curve below can be found as
follows: the area between 0 and -1=0.3413, this is the same as the area between 0 and
1; and the area between 0 and 0.50=0.1915;

The information is summarized in the following diagram:

b. What proportion of students has IQ above 140?

P(x>140) =P (z>1)=0.5-0.3413=0.1587 Note: z= (140-120)/20=1; the area

between 0 and 1 is 0.3413, and the area of the right half of the curve is 0.5, to find the
shaded area below, we find the difference between 0.5 and 0.3413.
c. What proportion of students has IQ below 150?

P(x<150)= P(z<1.50)=0.5+0.4332=0.9332 Note: z=(150-120)/20=1.50; the

area below 1.50 is obtained by summing the area between 0 and 1.50 i.e.
0.4332 and the area of the entire left half i.e. 0.5.

d. What proportion of students has IQ between 140 & 150?

P(140<x<150)=P(1.00<z<1.50)=0.4332-0.3413=0.0919

Note: the values are already converted to the z form above, to find the area
under question in the curve below, we find the difference between (the area
between 0 and 1.50, i.e. 0.4332), and (area between 0 and 1.00, i.e. 0.3413)
e. If there are 50 examinees, how many will have IQ score below 100 approximately?

P(x<100)=P(z<-1)= 0.5-0.3413=0.1587
0.1587 x 50≈ 8 examinees
Note: the area between 0 and 1 is 0.3413, and the left half of the curve is 0.5;
thus the total shaded area is given by the sum of 0.3413 and 0.5. To find the
number of examinees having IQ below 100, we multiplied the probability by the
size of the examinees.

f. What proportion of students has IQ between 120 and 200,

P(120<x<200)=P(0<z<4)=almost 0.5

Note: as 120 is the mean, its value in the z form is 0; and z=(140-120)/20=4; the area
designated by the range stated above covers the lion share of the right half of the
normal curve, the answer is almost half; we cannot say 0.5 as the indicated area goes
as far as z value of only 4.00, it is known that the z value extend infinitely on both
sides.
g. Practically almost all of the IQ score (99.72%) is between which two scores?
µ±3σ=120±3*20=120±60=60-180

Exercise 17

1. The time needed to complete a final examination in a particular college course is

normally distributed with a mean of 80 minutes and a standard deviation of 10
minutes. Answer the following questions.
a. What is the probability of completing the exam in one hour or less?
b. What is the probability that a student will complete the exam between 60
minutes and 75 minutes?
c. Assume that the class has 60 students and the examination period is 90
minutes in length. How many students do you expect will be unable to
complete the exam in the allotted time?

2. What Service life of truck tyres for heavy-duty trucks follows the normal
distribution with mean 50000 km and standard deviation 5000 km;
a. Find the probability that a tyre will last:
i. between 47,000 km and 60000 km
ii. at least 48500 km
iii. more than 48,000 km
iv. at most 60,000 km
v. above 70,000 km
b. What are the median and modal service life kilometers of tyres?

c. Is the distribution of the service life kilometers symmetrical?

Synopsis

Probability for any range of values of normal random variable can be found from the
standard normal (z) distribution table, convert values using the following formula

Wrap up Discussion Questions:

 How is probability found for any range of values of normal random variable from
the z distribution (table)?

Next Session’s Assignment:

 Attempt exercise 17, #1 & 2

 Read about finding unknown value of normal random variable, given probability.

Topic 18: Continuous Probability Distribution: Finding Values of

Unknown Normal Random Variable, Given a Probability

Topic Learning Objectives:

By the end of this session students are expected to:

 Find the corresponding value of unknown z or x when an area under the

normal distribution curve is known;

Topic outline

1. Finding Values of Unknown Normal Random Variable, given a probability

2. Synopsis
3. Wrap up discussion questions
4. Next session’s assignment
Reading Assignment Discussion:

 How should the normal probability be used to find unknown value of x given
probability (area)?
Reading Text:

So far, we have been finding probability for a given range of value (s) of normal
random variable, by converting or transforming x to z; conversely, in this section, the
inverse transformation of z to x, given probability, will be discussed. A further
application of the normal distribution involves finding the value of the observation x
when the area or probability (percent) above or below the observation is given. To find
the unknown x (value of normal random variable) follow the steps below:
 Draw pictures of the normal distribution in question and of the standard normal
distribution.
 Shade the area corresponding to the desired probability.
 From the table of the standard normal distribution, find the z value or values
corresponding to the area or probability.
 Use the inverse transformation from z to x to get value(s) of the original random
variable as follows:

Example 18

In Wollega University, an analysis of the final test scores for Managerial Statistics
reveals the scores follow the normal probability distribution. The mean of the
distribution is 75 and the standard deviation is 8.
Required:
a. If the Instructor wants to award an A to students whose score is in the highest
10%. Students will get A above which score?
b. Find the limits of the central/average 50% of scores obtained by the students.
 c. If students with the least 2% of the score will be given F, below which score is F
assigned?
d. 70% of the score is below which point?
e. 95.44% of the score is between which two points?
Solution: Given: µ=75, σ=8;
a. Area (probability)=10%=0.1; To get the corresponding z value from the table:
0.5-0.1=0.4000; z0.4000=1.28; pictorial representation of the problem is given
below:

Z=
X=?
X=µ+zσ=75+1.28*8=75+10.24=85.24
b. Area (probability)=0.5 at the center; z0.25=±0.67

X=µ+zσ=75±0.67*8=75±5.36=69.64-80.36
 c. Area (probability)=0.02; 0.5-0.02=0.4800; z0.4800=-2.05

0.5-0.02=0.4800

0.02

z=-2.05 0
X=µ+zσ=75-2.05*8=75-16.4=58.6
d. Area (probability)=0.7 (0.5+0.2); z0.2000=0.52

X=µ+zσ=75+0.52*8=75+4.16=79.16
e. For a normal distribution 95.44% of the distribution is within µ±2σ; thus,
75±2*8=75±16=the scores are between 59 and 91

Exercise 18

1. The average age of a person married for the first time is 26 years. Assume the
ages of the first marriages have a normal distribution with a standard deviation
of 4 years.
a. Before what age do 90% of people get married for the first time?
b. If the 20% of people getting married for the first time very late are given
a counseling service by a certain NGO what is the least age that deserves
the service?
 c. Find the limits of ages that the central/average 50% of people getting
married for the first time has.
d. 5% of people making marriage at the earliest age, for the first time, are
found below which age?
e. Practically almost all people getting married for the first time do so
between which two ages?
2. Individuals adopt new products or innovation or idea at different time and rate.
People can be classified into 5 adopter categories as shown in Figure 18.1. As
shown by the normal curve, after a slow start, an increasing number of people
adopt the new product. The number of adopters reaches a peak and then drops
off as fewer non-adopters remain. As successive groups of consumers adopt the
innovation (the rising curve), it eventually reaches its saturation level. Innovators
are defined as the first 2.5 percent of buyers to adopt a new idea; the early
adopters are the next 13.5 percent; and so forth. Suppose the mean adoption
time for a new product is 1800 days with a standard deviation of 200 days,
answer the following based on Figure 18.1.
a. Innovators will adopt the new product as late as how many days?
b. Laggards will adopt the new product after how many days?
c. The middle majority (early majority and late majority together) are
expected to adopt the new product within upper and lower limits of how
many days?
d. Early adopters in the population are expected to start adopting the new
product between which two days?
e. Indicate how many standard deviations are each group’s adoption time is
from the mean adoption time?
f. 68.26% of the population adopt the new product between which two
days?
 A
B C D E

A. B. C. D. E.
Adoption time

Figure 18.1 Time of Adoption of Innovation

Synopsis
 We use the inverse transformation formula, when we want to get from a given
probability the value or values of a normal random variable X as follows:
o

Wrap up Discussion Questions:

 Explain how the inverse transformation is used to find unknown value of random
variable (x) given probability.

Next Session’s Assignment:

 Attempt exercise 18, #1 & 2

 Read about finding unknown mean and standard deviation for a normal
distribution

Topic 19: Continuous Probability Distribution: Finding Unknown Mean

and Standard Deviation of a Normal Probability Distribution

Topic Learning Objectives:

By the end of this session students are expected to:

 Find unknown mean and standard deviation for a normal distribution given
two arbitrary values and corresponding area under the normal curve
Topic outline

1. Finding Unknown Mean and Standard Deviation of a Normal Probability

Distribution
2. Synopsis
3. Wrap up discussion questions
4. Next session’s assignment

Reading Assignment Discussion:

 How should unknown mean and standard deviation be found given values of
x, and probabilities?
Reading Text:

Sometimes the mean and the standard deviation of a normal distribution may not be
given. In such situations the given probability of the two values of normal random
variables (x1 and x2) are used to compute the mean and the standard deviation. From
the given probabilities find the corresponding z values and formulate the following pair
of equations and solve them simultaneously and find the unknown mean and standard

deviation:

N.B.: the mean and standard deviation are the same for the pair of equations because
the values and probabilities belong to the same normal distribution.

Example 19

The heights of soldiers are normally distributed. If 11.51% of the soldiers are taller than
70.4 inches and 9.68% are shorter than 65.4 inches, find the mean & standard
deviation for the data of heights of soldiers.

From area (probability) of 0.1151 on the right extreme of a normal curve, 0.5-
0.1151=0.3849; Z0.3849=1.20;
From area (probability) of 0.0968 on the left extreme of a normal curve, 0.5-
0.0968=0.4032, Z0.4032=-1.30

= Thus, σ=2 inches and µ=68 inches

Exercise 19

1. Construction time for a certain building is normally distributed with mean &
unknown variance; we do know that 75% of the time construction takes within a
maximum of 12 month. 45% of the time construction takes within at most 10
month. Find the mean & Standard deviation
2. The annual sales are normally distributed with unknown mean and unknown
standard deviation. 40% of the time sales are more than 480,000 and 10% of
the time sales are more than 500,000. What are the mean and the standard
deviation?

Synopsis
 Unknown mean and standard deviation can be found for a normal probability
distribution given 2 values of a normal random variable and related probabilities
solving the following simultaneously:

Wrap up Discussion Questions:

 Explain how unknown mean and standard deviation could be found given
probabilities and related values of normal random variable.

Next Session’s Assignment:

 Attempt exercise 19 #1 & 2

 Read about normal approximation of a binomial and Poisson probabilities or
distribution
Topic 20: Continuous Probability Distribution: Normal Approximations
of Binomial & Poisson Probabilities

Topic Learning Objectives:

By the end of this session students are expected to:

 Approximate the binomial probability distribution using the standard normal

probability distribution
 Approximate the Poisson probability distribution using the standard normal
probability distribution

Topic outline

1. Normal Approximations of Binomial & Poisson Probabilities

2. Approximating Poisson Distribution using Normal Probability Distribution
3. Synopsis
4. Wrap up discussion questions
5. Next session’s assignment

Reading Assignment Discussion:

 State the conditions when the binomial and Poisson probability distributions
are better approximated by the normal probability distribution

Reading Text:

Approximating Binomial Distribution using Normal Probability Distribution

As n gets big using tables or formula to generate the corresponding binomial

probabilities may not be convenient. Consequently, the more efficient approach is to
use normal approximation to the binomial. We can use the normal distribution (a
continuous distribution) as a substitute for a binomial distribution (a discrete
distribution) for large values of n because, as n increases, a binomial distribution gets
closer and closer to a normal distribution.
Rule: Good approximation of the binomial distribution by the normal distribution occurs
when np≥5 & nq ≥5 and n >20.

Since continuous variables exist in intervals continuity correction factor is required to

change the discrete variable to continuous variable.

Rule: Suppose ‘a’ represents certain discrete value; adjustment using the continuity
correction factor is made as follows:
 for x > a, add 0.5; thus it will be x > a+0.5
 for x ≤ a, add 0.5; thus it will be: or x ≤ a+0.5
 for x < a, subtract 0.5; thus it will be x< a-0.5
 for x ≥ a, subtract 0.5; thus it will be x ≥ a-0.5

Using the binomial approach, find the mean (µ)=n.p and standard deviation
(σ)= ; finally, find the z score and compute probability.

Example 20

1. It is believed that 30% of all stolen cars are recovered and returned to the
owners. In a month when 200 cars are stolen,
a. What is the expected number of cars that are recovered and returned seats?
µ=n.p=200*0.3=60
b. What is the variance and standard deviation of the cars that are recovered
and returned?
Variance=n.p.q=200*0.3*0.7=42 S.dev.= =6.48
c. What is the probability (applying normal approximation to the
binomial) that:
i. Fewer than 65 cars will be recovered and returned to their owners?
Note: n=200, P=0.3, Q=0.7; thus n>20 and np & nq≥5 is satisfied and
the binomial problem can be approximated using normal distribution
 P(x<65); adjust using continuity correction factor (subtract 0.5); P(x<65-
0.5)= P(x<64.5); convert to z; Z= 64.5-60/6.48=0.69
P(z<0.69)=0.5-0.2549=0.2451
ii. More than 75 cars will be recovered and returned to their owners?
P(x>70); adjust P(x>70+0.5)=P(x>70.5); z=75.5-60/6.48=2.39
P(z>2.39)=0.5-0.4916=0.0084
iii. Exactly 70 cars recovered and returned to their owners?
P(x=70) is the same as P(x≥70 and x≤70); adjust P(x≥70-0.5 and
x≤70+0.5)
P(69.5≤x≤70.5)=P(1.47≤z≤1.62)=0.4474-0.4292=0.0182

Approximating Poisson Distribution using Normal Probability Distribution

When the mean of a Poisson is relatively large, the normal prob. dist. can be used to
approximate it. Rule: Good approximation is when µ≥10

2. The average number of calls for a service received by a machine repairer shop
per 8hour shift is 10. What is the probability that more than 15 calls will be
received during a randomly selected 8 hours shift?
µ=10, for Poisson, Standard deviation= = =3.16;
P(x>15)= adjust; P(x>15.5); convert to z, z=15.5-10/3.16)=
P(z>1.74)=0.5-0.4591=0.0409
Exercise 20
1. For an Airline 80% of the time seats in all flights are occupied. If a particular Air
plane has 100 seats:
a. What is the expected number of occupied seats?
b. What is the variance and standard deviation of the occupied seats?
c. What is the probability (applying normal approximation to the binomial)
that:
i. More than 85 seats will be occupied
ii. At most 88 seats will be occupied
 iii. Between 76 and 82 seats will be occupied
iv. Exactly 88 seats will be occupied
2. Patients arrive at a hospital at an average rate of 25 per day. What is the probability
that more than 22 patients will arrive in a day? Assuming arrival follow Poisson
distribution.

Synopsis
 Binomial probability distribution can best be approximated by the normal
probability distribution when np≥5 and nq≥5; and n>20
 Poisson probability distribution can best be approximated by the normal
probability distribution when µ ≥10
o Continuity correction factor (adding or subtracting 0.5) is used to adjust
discrete values into continuous forms
o Using the mean and standard deviation of each distributions; z value is set
,and hence, the probability

Wrap up Discussion Questions:

 Explain how the continuity correction factor is used

 What are the conditions that must be satisfied to better approximate binomial
and Poisson distributions using the normal distribution?

Next Session’s Assignment:

 Attempt exercise 20 #1 & 2

 Read about basic concepts of sampling