Probability

Distribution
 Definitions
• Trial: A single performance of a well-defined experiment.
• Bernoulli Trial: An experiment is called Bernoulli trial if it
has two possible outcomes namely success and failure.
• (conditions for Binomial Distribution) Binomial
Experiment: An experiment is called Binomial
experiment when it has the following properties:
– The experiment consists of n repeated trials
– Each trial has two possible outcomes namely success and
failure.
– The probability of success, p remains the same from trial to
trail.
– The repeated trials are independent.
 Binomial Distribution
• Let A be some event associated with a random experiment E,
such that 𝑃(𝐴) = 𝑝 and 𝑃 𝐴ҧ = 1 − 𝑝 = 𝑞. Assuming that 𝑝
remains the same for all repetitions, if we consider 𝑛
independent repetitions (or trials) of 𝐸 and if the random
variable (RV) 𝑋 denotes the number of times the event A has
occurred, then X is called a binomial random variable with
parameters n and p or we say that 𝑋 follows a binomial
distribution with parameters n and p, or symbolically
𝐵(𝑛, 𝑝). Denoted by 𝑋~𝐵(𝑥; 𝑛, 𝑝). Obviously, the possible values
that 𝑋 can take, are 0, 1, 2, … , 𝑛.
Then, the probability mass function of the binomial random
variable X is given by
𝑛 𝑥 𝑛−𝑥
𝑃 𝑋=𝑥 =𝑝 𝑥 = 𝑝 𝑞 ; 𝑥 = 0,1,2, … , 𝑛
𝑥
Where, 𝑝 + 𝑞 = 1
 Mean of Binomial Distribution
𝑀𝑒𝑎𝑛, 𝐸 𝑋 = ∑𝑥𝑝 𝑥
𝑛
𝑛 𝑥 𝑛−𝑥
= ෍𝑥 𝑝 𝑞
𝑥
𝑥=0
𝑛 0 𝑛 𝑛 𝑛 2 𝑛−2
=0⋅ 𝑝 𝑞 +1⋅ 𝑝 ⋅ 𝑞 𝑛−1 + 2 ⋅ 𝑝 𝑞 + ⋯ + 𝑛𝑝𝑛
0 1 2
𝑛 − 1 𝑛−2 𝑛−1
= 𝑛𝑝[𝑞 𝑛−1 + 𝑞 𝑝+ 𝑞 𝑛−3 𝑝2 + ⋯ + 𝑝𝑛−1 ]
1 2
𝑛−1
= 𝑛𝑝 𝑞 + 𝑝
= 𝑛𝑝 [∴ 𝑝 + 𝑞 = 1]
• Variance = 𝑛𝑝𝑞
• Standard Deviation= 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = 𝑛𝑝𝑞
 Binomial Distribution
• Properties:
– It is a discrete distribution with parameters n and p.
– The mean of the distribution is 𝑛𝑝 and its variance is
𝑛𝑝𝑞. The mean of the distribution is greater than
variance since 𝑞 < 1.
1
– The distribution is positively skewed if 𝑝 < and
1 2
negatively skewed if 𝑝 > .
2
1
– The distribution is symmetric if 𝑝 = 𝑞 =
2
– The distribution tends to Poisson distribution if the
number of trials, n tends to infinity.
– The distribution tends to normal distribution if n tends
to infinity and p or q are not so small.
 Practical Examples
• Number of defective bolts in a box containing n
bolts
• Number of post graduates in a group of n
people.
• Number of oil wells yielding natural gas in a
group of n wells test drilled.
• Number of machines lying idle in a factory
having n machines.
• Number of workers suffer from occupational
disease in a randomly selected sample of 10
workers.
 Example
• If 10% of the screws produced by a machine are defective, find the probability
that out of 5 screws chosen at random,
i. None is defective,
ii. One is defective and
iii. At most two is defective.
• Solution: Probability that x screws out of 5 screws are defective
𝑛 𝑥 𝑛−𝑥 5
𝑃 𝑋=𝑥 = 𝑝 𝑞 = (0.1)𝑥 (0.905−𝑥 ,
𝑥 𝑥
𝑤ℎ𝑒𝑟𝑒, 𝑥 = 0,1,2, … , 5
i. Probability that none of the screws out of 5 screws is defective, 𝑃 𝑋 = 0 =
5
0.1 0 0.9 5 = 0.5905
0
5 1 4
ii. Probability that one screw out of 5 screws is defective 𝑃 𝑋 = 1 = 0.1 0.9 =
1
0.3281
iii. Probability that at most 2 screws out of 5 screws are defective
𝑃 𝑋 ≤2 =𝑃 𝑋 =0 +𝑃 𝑋 =1 +𝑃 𝑋 =2
= 0.9914
 Exercise
• It has been claimed that in 60% of all solar heat installations the utility bill is reduced by at least
one-third. Accordingly, what are the probabilities that the utility bill will be reduced by at least
one third in
– Four of five installations
– At least four of five installations.
• The incidence of an occupational disease in an industry is such that the workers have a 20%
chance of suffering from it. What is the probability that out of 6 workers chosen at random,
four or more will suffer from the disease.
• In a multiple-choice examination, there are 20 questions. Each question has 4 alternative
answers following it and the student must select one correct answer. 4 marks are given for a
correct answer and 1 mark is deducted for a wrong answer. A student must secure at least 50%
of the maximum possible marks to pass the examination. Suppose a student has not studied at
all, so that he answers the questions by guessing only. What is the probability that he will pass
the examination?
• The probability of a man hitting a target is 1 . 3 (i) If he fires 5 times, what is the probability of
his hitting the target at least twice? (ii) How many times must he fire so that the probability of
his hitting the target at least once is more than 90%?
• In sampling a large number of parts manufactured by a machine, the mean number of
defectives in a sample of 20 is 2. Out of 1000 such samples, how many would be expected to
contain exactly two defective parts?
 Poisson Distribution
A discrete random variable X is said to have a Poisson distribution if
its probability function is defined by
𝑒 −𝜆 𝜆𝑥
𝑝 𝑥 = ; 𝑥 = 0,1,2, … , ∞
𝑥!
Where, 𝑒 = 2.71828 and 𝜆 is the only parameter of the distribution
which is the mean of the distribution and 𝑥 is any positive value that
X can assume. This distribution is denoted by, 𝑋~𝑃 𝑥; 𝜆
 Properties
• The mean and variance of the distribution is
𝜆. So, the mean and variance of the
distribution are equal.
• When 𝜆 tends to ∞, Poisson distribution
tends to normal distribution.
• The distribution is positively skewed and
leptokurtic.
 Conditions
• The events occur independently of each other.
• The events occur singly and at a random in continuous
space or time.
• The events occur at a constant rate in this sense that the
mean number of occurrences in the given space or
interval of time is proportional to the lengths of space or
time.
• The random variable 𝑋 should be discrete.
• The numbers of trials 𝑛 is very large.
• The probability of success 𝑝 is very small (very close to
zero).
• 𝜆 = 𝑛𝑝 is finite.
• The occurrences are rare.
 Practical Examples
• Number of defective bulbs produced by a
reputed company
• Number of telephone calls per minute at a
switchboard
• Number of cars passing a certain point in one
minute
• Number of printing mistakes per page in a large
text
• Number of persons born blind per year in a
large city
 Limiting case
• Poisson distribution is a limiting case of
binomial distribution under the following
conditions:
– The number of trials should be infinitely large,
i.e., 𝑛 → ∞.
– The probability of successes p for each trial
should be very small, i.e., 𝑝 → 0.
– 𝑛𝑝 = 𝜆 should be finite where 𝜆 is a constant.
Constants of Poisson
Distribution
 Example
• A car hire firm has two cars, which hires out
by the day. The number of demands for a car
on each day is Poisson distributed with
parameter 1.5. Calculate
– the proportion of days on which neither of the
cars is used
– proportion of days on which some demand
cannot be met for lack of cars or is refused.
[Given 𝑒 −1.5 = 0.223 ]
 Exercise
• In a certain factory turning razor blades, there is a small chance of 1/500 for any blade to be
defective. The blades are in packets of 10. Use Poisson distribution to calculate the approximate
number of packet containing.
– No defective
– one defective and
– two defective blades respectively in one consignment of 10,000 packets
• If 2% of lightbulbs are defective, find the probability that
– at least one is defective, and
– exactly 7 are defective. Also, find P(1 < X < 8) in a sample of 100.
• Suppose a book of 585 pages contains 43 typographical errors. If these errors are randomly
distributed throughout the book, what is the probability that 10 pages, selected at random, will
be free from errors?
• A hospital switchboard receives an average of 4 emergency calls in a 10-minute interval. What is
the probability that (i) there are at most 2 emergency calls? (ii) there are exactly 3 emergency
calls in an interval of 10 minutes?
• A manufacturer, who produces medicine bottles, finds that 0.1% of the bottles are defective.
The bottles are packed in boxes containing 500 bottles. A drug manufacturer buys 100 boxes
from the producer of bottles. Using Poisson distribution, find how many boxes will contain (i) no
defective bottles and (ii) at least 2 defective bottles.
• The number of accidents in a year attributed to taxi drivers in a city follows Poisson distribution
with a mean of 3. Out of 1000 taxi drivers, find approximately the number of drivers with (i) no
accidents in a year, and (ii) more than 3 accidents in a year
 Importance of Poisson Distribution
• Utility or Importance of Poisson Distribution. The conditions under which Poisson distribution is obtained as a
limiting case of the Binomial distribution and also the conditions for the general model underlying Poisson
distribution [c.f. remark 5] suggest that Poisson distribution can be used to explain the behaviour of the
discrete random variables where the probability of occurrence of the event is very small and the total number
of possible cases is sufficiently large. As such Poisson distribution has found application in a variety of fields
such as Queuing Theory (waiting time problems), Insurance, Physics, Biology, Business, Economics and
Industry. Most of the Temporal Distributions (dealing with events which are supposed to occur in equal
intervals of time) and the Spatial Distributions (dealing with events which are supposed to occur in intervals
of equal length along a straight line) follow the Poisson Probability Law. We give below some practical
situations where Poisson distribution can be used :
• The number of telephone calls arriving at a telephone switch board in unit time (say, per minute).
• The number of customers arriving at the super market ; say per hour.
• The number of defects per unit of manufactured product [This is done for the construction of control chart for
number of defects (c) in Industrial Quality Control].
• To count the number of radio-active disintegrations of a radio-active element per unit of time (Physics).
• To count the number of bacteria per unit (Biology).
• The number of defective material say, pins, blades etc. in a packing manufactured by a good concern.
• The number of suicides reported in a particular day or the number of causalities (persons dying) due to a rare
disease such as heart attack or cancer or snake bite in a year.
• The number of accidents taking place per day on a busy road.
• The number of typographical errors per page in a typed material or the number of printing mistakes per page
in a book.
Negative Binomial distribution
• Properties:
– The experiment consists of a series of
independent and identical Bernoulli trials, each
with probability p of success.
– The trials are observed until exactly r successes
are obtained, where r is fixed by the
experimenter.
– The random variable X is the number of trials
needed to obtain r successes.
Negative Binomial Distribution
Let X denotes the number of trials needed to
produce r successes. The Binomial random
variable X assumes a value x, on which the 𝑟𝑡ℎ
success occurs, then the number of trials is
given by
𝑥−1
𝑃{𝑋 = 𝑥} = (1 − 𝑝)𝑥−𝑟 𝑝𝑟
𝑟−1
𝐻𝑒𝑟𝑒, 𝑥 = 𝑟, 𝑟 + 1, 𝑟 + 2, …
Where, 𝑝 and 𝑞 have the usual meaning in
Bernoulli's trials.
Negative Binomial Distribution
𝑛𝑞
• Mean=
𝑝
𝑛𝑞
• Variance= 2
𝑝
 Example
• If the probability is 0.40 that a child exposed
to a certain disease, will it contain it. What is
the probability that the tenth child exposed
to the disease will be the third to catch it?
• Solution: Required probability,
9 𝑟 𝑥−𝑟
𝑃 𝑋 = 10 = 𝑝 𝑞
2
9
= 0.4 3 0.6 7
2
= 0.0645
 Example
An item is produced in large numbers. The machine is known to produce 2%
defectives. A quality control inspector is examining the items by taking them one by
one at random. What is the probability that at least 4 items are to be examined in
order to get 2 defectives?
• Solution: Here, success, 𝑝 = 0.02
failure, 𝑞 = 0.98
If at least 4 items are to be examined (viz. 4 trails are required) to give 2 defectives
(success), 4 or 5 or 6 or … items are to be examined i.e., the first 3 or 4 or 5 or …
trails must result in 1 success and the next trail in a success.
Required probability
3 2 2 4 2 3 5 2 4
= 𝑝 𝑞 + 𝑝 𝑞 + 𝑝 𝑞 +⋯
1 1 1
= 𝑝2 3𝑞 2 + 4𝑞 3 + 5𝑞 4 + ⋯
= 𝑝2 1 − 𝑞 −2 − 1 − 2𝑞
= 1 − 𝑝2 − 2𝑝2 𝑞
= 1 − 0.02 2 − 2 × 0.02 2 × 0.98
= 0.9988
 Geometric Distribution
• Properties:
– The experiment consists of a series of trials. The
outcome of each trial can be classed as being either
a success or a failure i.e., trial is a Bernoulli trial
– The trials are identical and independent in the
sense that outcome of one trial has no effect on the
outcome of any other.
– The probability of success , p, remains the same
from trial to trial.
– The random variable X denotes the number of trials
needed to obtain the first success
 Geometric Distribution
Sample space of the distribution looks like
𝑆 = {𝑠, 𝑓𝑠, 𝑓𝑓𝑠, 𝑓𝑓𝑓𝑠, … }
• Here the random variable X denotes the number of trials needed to obtain the
first success. Certainly, X assumes the values 1,2,3, ⋯ .Thus,
𝑃 𝑋 = 1 = 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑜𝑛 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 𝑡𝑟𝑖𝑎𝑙 = 𝑝
𝑃 𝑋 = 2 = 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 𝑜𝑛 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 𝑡𝑟𝑖𝑎𝑙 𝑎𝑛𝑑 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑜𝑛 𝑡ℎ𝑒 𝑠𝑒𝑐𝑜𝑛𝑑 𝑡𝑟𝑖𝑎𝑙
= 𝑃 𝑓𝑎𝑖𝑙 𝑜𝑛 𝑓𝑖𝑟𝑠𝑡 𝑡𝑟𝑖𝑎𝑙 × 𝑃[𝑠𝑢𝑐𝑐𝑒𝑑𝑒𝑑 𝑜𝑛 𝑡ℎ𝑒 𝑠𝑒𝑐𝑜𝑛𝑑 𝑡𝑟𝑖𝑎𝑙]
= 1−𝑝 ×𝑝
= 𝑝𝑞 2−1
𝑃 𝑋=3
= 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 𝑜𝑛 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 𝑎𝑛𝑑 𝑠𝑒𝑐𝑜𝑛𝑑 𝑡𝑟𝑖𝑎𝑙 𝑎𝑛𝑑 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑜𝑛 𝑡ℎ𝑒 𝑡ℎ𝑖𝑟𝑑 𝑡𝑟𝑖𝑎𝑙
= 1−𝑝 1−𝑝 𝑝
= 𝑝𝑞 3−1
Continuing the process,
𝑃 𝑋 = 𝑥 = 𝑝𝑞 𝑥−1 , 𝑥 = 1,2,3, …
 Geometric Distribution
Let the Random variable X denote the number of trials of a random
experiment required to obtain the first success (occurrence of an
event A). Obviously, X can assume the values 1, 2, 3, …
Now, 𝑋 = 𝑟, if and only if the first (𝑟 – 1) trials result in failure
(occurrence of 𝐴) and the 𝑟𝑡ℎ trial results in success (occurrence A).
Hence,
𝑃(𝑋 = 𝑟) = 𝑞 𝑟−1 𝑝; 𝑟 = 1, 2, 3, … , ∞
where 𝑃 𝐴 = 𝑝 and 𝑃 𝐴ҧ = 𝑞.
If 𝑋 is a discrete random variable that can assume the values
1, 2, 3, … , ∞ such that its probability mass function is given by
𝑃(𝑋 = 𝑟) = 𝑞 𝑟−1 𝑝; 𝑟 = 1, 2, . . , ∞
where 𝑝 + 𝑞 = 1
then X is said to follow a geometric distribution.
 Geometric Distribution
1
• Mean=
𝑝
𝑞
• Variance= 2
𝑝
 Example
• If the probability that an applicant for a driver’s license will pass
the road test on any given trial is 0.8, what is the probability that
he will finally pass the test (a) on the fourth trial, and (b) in fewer
than 4 trials?
• Solution: Let X denote the number of trials required to achieve
the first success. Then X follows a geometric distribution given by
𝑃 𝑋 = 𝑟 = 𝑞 𝑟−1 𝑝; 𝑟 = 1,2,3, … , ∞
Here, 𝑝 = 0.8 𝑎𝑛𝑑 𝑞 = 0.2
❑ 𝑃 𝑋 = 4 = 0.8 × 0.2 4−1
= 0.0064
❑ 𝑃 𝑋 < 4 = ∑3𝑟=1 0.8 × 0.2 𝑟−1
= 0.8 0.2 0 + 0.2 1 + 0.2 2

= 0.9984
 Exercise
• A coin is tossed until the first head occurs. Assuming that the
tosses are independent and the probability of a head occurring is
p, find the value of p so that the probability that an odd number
of tosses is required is equal to 0.6. Can you find a value of p so
that the probability is 0.5 that an odd number of tosses is
required?
• In a certain manufacturing process, it is known that, on the
average, 1 in every 100 items is defective. What is the probability
that the fifth item inspected is the first defective item found?
• At "busy time" a telephone exchange is very near capacity, so
callers have difficulty placing their calls. It may be of interest to
know the number of attempts necessary in order to gain a
connection. Suppose that we let p = 0.05 be the probability of a
connection during busy time. We are interested in knowing the
probability that 5 attempts are necessary for a successful call.