COURSE TITLE: Linear Algebra, Statistics and

Probability
COURSE CODE: MATH-2101

Instructor: M. Ershadul Haque

Associate Professor

Department of Statistics, DU
 The Binomial Distribution

• The binomial distribution is one of the most widely used probability distributions in applied
statistics.

• The distribution is derived from a process known as a Bernoulli trial,

• When a random process or experiment, called a trial, can result in only one of two mutually
exclusive outcomes, such as dead or alive, sick or well, full-term or premature, the trial is
called a Bernoulli trial.
 The Binomial Distribution (Cont…)

• Binomial Random Variable and Its pmf: If a Bernoulli trial can result in a success with
probability 𝑝, then the probability distribution of the random variable 𝑋, the number of
successes in 𝑛 independent trials, is called binomial distribution with probability function

𝑓 𝑥; 𝑛, 𝑝 = 𝑛𝐶𝑥 𝑝 𝑥 1 − 𝑝 𝑛−𝑥 , 𝑥 = 0, 1, 2, … , 𝑛, 0 < 𝑝 < 1.

• The requirements for using the binomial distribution are as follows:

 There are only two possible outcomes: success or failure in a single trial.

 The number of trials, 𝑛, must be fixed, regardless of the outcome of each trial.

 The trials are independent.

 The probability of success, 𝑝, in each trial is constant.

• Suppose that 𝑋 follows binomial distribution with number of trials 𝑛 and success probability
𝑝. Notationally it is represented by 𝑋 ∼Bin(𝑛, 𝑝).
 The Binomial Distribution (Cont…)

• Note: If we put 𝒏 = 𝟏, the resulting distribution is Bernoulli distribution.

• The mean and variance of binomial random variable are

Mean: 𝜇 = 𝐸 𝑋 = 𝑛𝑝, 𝜎 2 = 𝑉 𝑋 = 𝑛𝑝 1 − 𝑝

• The mean and variance of Bernoulli random variable are

Mean: 𝜇 = 𝐸 𝑋 = 𝑝, 𝜎 2 = 𝑉 𝑋 = 𝑝 1 − 𝑝
Note: If 𝑋 ∼Bin 𝑛, 𝑝 then 𝜇 > 𝜎 2
 The Binomial Distribution (Cont…)

• Problem: Suppose that 10 percent of a certain population is color blind. If a random sample of
10 people is drawn from this population, find the probability that (a) Two or fewer will be color
blind. (b) Two or more will be color blind. (c) Between three and five inclusive will be color blind.

• Solution: (i) Total number of trials: 𝑛 = 10, and probability of success 𝑝 = 0.10. Let 𝑋 be the
number of color blind population in the sample, then 𝑋 have binomial distribution with pmf
𝑃 𝑋 = 𝑥 = 𝑛𝐶𝑥 𝑝 𝑥 1 − 𝑝 𝑛−𝑥 = 10𝐶𝑥 0.1𝑥 0.910−𝑥 , 𝑥 = 0, 1, 2, … , 10.

(a) We have to find

𝑃 𝑋 ≤2 =𝑃 𝑋 =0 +𝑃 𝑋 =1 +𝑃 𝑋 =2
We have, 𝑃 𝑋 = 0 = 10𝐶0 0.10 1 − 0.1 10−0 = 1 × 1 × 0.3487 = 0.3487
Similarly 𝑃 𝑋 = 1 = 0.3874 and 𝑃 𝑋 = 2 = 0.1937
Therefore,𝑃 𝑋 ≤ 2 = 0.9298
(b) We have to find
𝑃 𝑋 ≥ 2 = 1 − 𝑃 𝑋 = 0 − 𝑃 𝑋 = 1 = 0.2639
(c) We have to find
𝑃 3 ≤ 𝑋 ≤ 5 = 𝑃 𝑋 = 3 + 𝑃 𝑋 = 4 + 𝑃 𝑋 = 5 = 0.0574 + 0.0112 + 0.0015 = 0.0701
 Binomial Distribution (cont…)

• Problem: Batches that consist of 24 manufacturing items from a production process are
checked for conformance to customer requirements. The mean number of nonconforming
items in a batch is 3. Assume that the number of nonconforming items in a batch, denoted
as 𝑋, is a binomial random variable. (i) What are 𝑛 and 𝑝? (ii) What is 𝑃 𝑋 ≥ 3 ?

• Solution: (i) Total number of trials: 𝑛 = 24

𝜇 = 𝑛𝑝 = 24 × 𝑝 = 3
3
So, probability of success, 𝑝 = = 0.125
24
We have to find
𝑃 𝑋 ≥3 =1−𝑃 𝑋 ≤2 =1−𝑃 𝑋 =0 −𝑃 𝑋 =1 −𝑃 𝑋 =2
We have, 𝑃 𝑋 = 0 = 24𝐶0 0.1250 1 − 0.125 24−0 = 24𝐶0 0.1250 0.87524 = 0.04057
Similarly 𝑃 𝑋 = 1 = 0.13909 and 𝑃 𝑋 = 2 = 0.22851
Therefore, 𝑃 𝑋 ≥ 3 = 1 − 0.40417 = 0.59183
 Poisson Distribution

• Poisson Distribution: Situations often arise where the variable of interest is the number of
occurrences of a particular event in a given interval of space or time. The events are often
defects, accidents or unusual natural happenings, such as earthquakes, where in theory
there is no upper limit on the number of events. Some examples are

 the number of cars passing a point on a road in a time interval of 1 minute

 the number of telephone calls received at a switchboard per minute

 the number of misprints on each page of a book,

 the number of radioactive particles emitted by a radioactive source in a time interval of

1 second.

 The number of plants infected with a particular disease in a plot of field.

 Poisson Distribution

• The Poisson distribution is often known as the distribution of rare events. Given an
interval of real numbers, assume counts occur at random throughout the interval. If the
interval can be partitioned into subintervals of small enough length such that

 For a small interval the probability of the event occurring is proportional to the size of
the interval.

 The probability of more than one occurrence in the small interval is negligible (i.e. they
are rare events). Events must not occur simultaneously

 Each occurrence must be independent of others and must be at random.

• The random experiment is called a Poisson process.

 The Poisson Distribution (cont…)

• Poisson distribution (definition): A random variable is said to have a Poisson distribution

with parameter 𝜆 > 0 if its pmf is given by

𝑒 −𝜆 𝜆𝑥
𝑓 𝑥; 𝜆 = , 𝑥 = 0, 1, 2, …
𝑥!

The Poisson distribution was first derived in 1837 by the French mathematician and physicist
Siméon-Denis Poisson (1781–1840).

• Suppose that 𝑋 follows Poisson distribution with average number of occurrence (mean) 𝜆.
Notationally it is represented by 𝑋 ∼Poi(𝜆).
 The Poisson Distribution (Cont…)

• Poisson approximation to Binomial distribution: If 𝑋 ∼Bin 𝑛, 𝑝 , and 𝑛 → ∞, 𝑝 → 0 such

that 𝑛𝑝 → 𝜆 (a constant) then it can be shown that 𝑓 𝑥; 𝑛, 𝑝 → 𝑓 𝑥; 𝜆 i.e., 𝑋 ∼Poi(𝜆)
 Poisson Distribution (cont…)

• Problem: The average number of spots in 1 square meter of metal sheet is 3.4. What is the
probability that there will be (a) 2 spots in the next square meter? (b) at most 2 spots in the
next square meter? (c) at least 3 spots in the next square meter?

𝑒 −𝜆 𝜆𝑥
• Solution: Given that 𝜆 = 3.4, we have 𝑓 𝑥; 𝜆 = , 𝑥 = 0, 1, 2, …
𝑥!

𝑒 −3.4 3.42
(a) 𝑃 𝑋 = 2 = = 0.1929
2!

(b) 𝑃 𝑋 ≤ 2 = 𝑃 𝑋 = 0 + 𝑃 𝑋 = 1 + 𝑃 𝑋 = 2 = 0.0334 + 0.1135 + 0.1929 = 0.3398

(c) 𝑃 𝑋 ≥ 3 = 1 − 𝑃 𝑋 ≤ 2 = 1 − 0.3398 = 0.6602

 Poisson Distribution (cont…)

• Problem: The average number of meteors found by a radar system in any one-minute
interval under specified conditions is 3.62. Assume the meteors appear randomly and
independently. (a) What is the probability that no meteors are found in a 30-second
interval? (b) What is the probability of observing at least four but not more than six meteors
in two minutes of observation? (c) Find the mean and standard deviation of meteors found
in three minutes interval.

• Solution: Given that 𝜆 = 3.62/min,

3.62 𝑒 −1.81 1.810

(a) we have 𝜆 = = 1.81 and 𝑃 𝑋 = 0 = = 0.164
2 0!

(b) Here 𝜆 = 3.62 × 2 = 7.24

𝑃 4 ≤ 𝑋 ≤ 6 = 𝑃 𝑋 = 4 + 𝑃 𝑋 = 5 + 𝑃 𝑋 = 6 = 0.082 + 0.119 + 0.143 = 0.344
(c) Here 𝜆 = 3.62 × 3 = 10.86
𝜇 = 𝜆 = 10.86, 𝜎 = 𝜆 = 10.86 = 3.3
 Poisson Distribution (cont…)

• Problem: Automobile battery of a particular brand malfunctions with probability 0.001. Use
two different distributions to calculate the probability that at least one out of 1000 batteries
will malfunction.

• Binomial distribution: Given that success probability , 𝑝 = 0.001 , let 𝑋: number of

batteries out of 1000 that will malfunction. Then 𝑋 ∼ Bin 1000,0.001 . Now we have
𝑃 𝑋 ≥ 1 = 1 − 𝑃 𝑋 = 0 = 1 − 0.3677 = 0.6323

• Poisson distribution: As 𝑛 → ∞, 𝑝 → 0 such that 𝑛𝑝 = 𝜆 = 1, then 𝑋 ∼Bin 𝑛, 𝑝 can be

approximated by 𝑋 ∼Poi(𝜆)

Now we have 𝑃 𝑋 ≥ 1 = 1 − 𝑃 𝑋 = 0 = 1 − 0.3679 = 0.6321

 Normal Distribution
• A continuous random variable 𝑋 is said to have a normal distribution if its pdf is given by

1 1 𝑥−𝜇 2
−
𝑓 𝑥; 𝜇, 𝜎 2 = 𝑒 2 𝜎 , −∞ < 𝑥 < ∞, −∞ < 𝑥 < ∞, 𝜎 > 0
𝜎 2𝜋

where 𝝁 and 𝝈𝟐 are two parameters of the distribution (𝝁 is the mean and 𝝈𝟐 is the variance)

• Suppose that 𝑋 follows normal distribution with mean and variance 𝝁 and 𝝈𝟐 , respectively.
Notationally it is represented by 𝑋 ∼ N(𝜇, 𝜎 2 ).

∞
• If ∼ N(𝜇, 𝜎 2 ), it can be shown that −∞
𝑓 𝑥; 𝜇, 𝜎 2 𝑑𝑥

• The mean and variance of normal random variable are

Mean: 𝐸 𝑋 = 𝜇, 𝑉𝑎𝑟 𝑋 = 𝜎 2

• The mean, median and mode of a normal random variable coincides.

 Normal Distribution (cont…)

• A normal distribution is a symmetrical distribution (symmetrical about the point 𝜇). The
normal distribution with mean 100 and variance 4 is shown in the following figure

90 95 100 105 110

 Normal Distribution (cont…)

• Difficulties in calculating probabilities: For an interval (𝑎, 𝑏) in real line, no analytical

2
𝑏 −1 𝑥−𝜇
solution for the following integrals 𝑎
𝑒 2 𝜎 𝑑𝑥 exist. However, the integral can be solved

by using numerical methods.

• The probability from normal distribution requires numerical methods.

• A table for calculating probabilities can be constructed using the numerical method. The
number of normal distributions is infinite in a normal family, each having a different mean
(𝜇), variance 𝜎 2 , or both. Thus providing these infinite numbers of normal tables is
impossible.

• Fortunately, one member of the normal family can be used to determine the probabilities for
all normal distributions. It is called the standard normal distribution.
 Standard Normal Distribution

• If 𝑋 is a normal random variable with mean 𝜇 and standard deviation 𝜎, then the random
𝑋−𝜇
variable 𝑍 = is called the standard normal variable and its probability distribution is
𝜎

called the standard normal distribution.

• The probability density function of standard normal variable is

1 1
− 𝑧2
𝑓 𝑧 = 𝑒 2 , −∞ < 𝑧 < ∞.
2𝜋

• The mean and variance of binomial random variable are

Mean: 𝐸 𝑍 = 0, 𝑉𝑎𝑟 𝑍 = 1
 Computing Probability for Normal Distribution

• IStep-1: Express 𝑃 𝑥1 < 𝑋 < 𝑥2 as 𝑃 𝑧1 < 𝑍 < 𝑧2

• Step-2: These probabilities for different intervals of a standard random variable (Z) can be
calculated by using a standard normal table.
 Problem on Normal Distribution
• Example: Suppose a random variable have a normal distribution with mean 283 and standard
deviation 1.65. Then 𝑃(𝑋 < 286) is equivalent to 𝑃 𝑍 < 1.82 , and 𝑃(𝑋 < 281) is equivalent to
𝑃 𝑍 < −1.21 as shown below
𝑋−𝜇 286−𝜇 286−283
𝑃 𝑋 < 286 = 𝑃 < =𝑃 𝑍< = 𝑃 𝑍 < 1.82 = F 1.82 = 0.9656
𝜎 𝜎 1.65
𝑋−𝜇 281−𝜇 281−283
𝑃 𝑋 < 281 = 𝑃 < =𝑃 𝑍< = 𝑃 𝑍 < −1.21 = F −1.21 = 0.1131
𝜎 𝜎 1.65
 Example (Cont…)
Similarly
𝑃 𝑋 > 286 = 𝑃 𝑍 > 1.82 = 1 − F 1.82 = 1 − 0.9656 = 0.0344
𝑃 𝑋 > 281 = 𝑃 𝑍 > −1.21 = 1 − F −1.21 = 1 − 0.1131 = 0.8869
 Example (Cont…)
Similarly
281.5 − 𝜇 𝑋 − 𝜇 285.4 − 𝜇
𝑃 281.5 < 𝑋 < 285.4 = 𝑃 < < = 𝑃 −0.91 < 𝑍 < 1.45
𝜎 𝜎 𝜎
= F 1.45 − F −0.91 = 0.9265 − 0.1814 = 0.7451
 Problem on Normal Distribution
• Problem: The weight of a sophisticated running shoe is normally distributed with a mean of 12
ounces and a standard deviation of 0.45 ounce. (a) What is the probability that a shoe weighs at most
12.2 ounces (b) What is the probability that a shoe weighs more than 13.2 ounces? (ii) What is the
probability that a shoe weighs is between 11.4 ounces and 12.9 ounces?

• Solution: Given that 𝜇 = 12, and 𝜎 = 0.5

𝑋−𝜇 12.2−𝜇 12.2−12

(a) 𝑃 𝑋 ≤ 12.2 = 𝑃 ≤ =𝑃 𝑍≤ = 𝑃 𝑍 ≤ 0.44 = 0.67
𝜎 𝜎 0.45
𝑋−𝜇 13.2−𝜇 13.2−12
(b) 𝑃 𝑋 ≥ 13.2 = 𝑃 ≥ =𝑃 𝑍≥ = 𝑃 𝑍 > 2.67 = 1 − 0.9962
𝜎 𝜎 0.45
11.4−𝜇 𝑋−𝜇 12.9−𝜇
(c) 𝑃 11.4 ≤ 𝑋 ≤ 12.9 = 𝑃 ≤ ≤ =𝑃 −1.33 ≤ 𝑍 ≤ 2 = 0.9772 − 0.0918
𝜎 𝜎 𝜎
Thank You