PROBABILITY DISTRIBUTIONS

Case Study at Sam Efromovich & Co.

Example (Application in Market Research): Sam Efromovich & Co
is a plant-based meat company which offers a range of plant-based
meat products. Market research at Sam provides the management
with up-to-date info on company’s existing products.

Given that the market share of Sam’s Vegan Meat is 40%, Sam’s
management wants to be the best in terms of taste and appearance
before coming up with a strategy for expanding to nonvegetarian
customers as well. Also, price is another factor which the management
needs to look at so as to not harm its existing market share.

So, the Market research team does a survey of a sample of 40 Vegan

Meat consumers to know details about the amount spent behing
Vegan Meat products, taste and appearance ratings of Sam’s Vegan
Meat and its competitors.
 Case Study at Sam Efromovich & Co.
Example (Application in Market Research):
It has been observed that on an average, a Vegan Meat consumer
spends about Rs. 500 per month and it may vary by Rs 100 from an
individual to the other.

Among other research questions, two questions may be

1. What is the probability that the number of customers preferring

Sam’s Vegan Meat in the sample be 10? Can we say anything
about its probability distribution?
2. What percentage of customers spend more than Rs. 650 per
month for buying Vegan meat products?
 Categorical Response Variables

Loan Approval Y=

Binary Response
Whether an Y=
employee left the
organization or not

Cause of Employee
Attrition Y=

Non-binary
4 Response
 Qualitative Variables
● Qualitative variables: Take unordered values
such as:
● Employee left the organization within one year:
{Yes, No}
● Email: {Spam, Not-spam}
● Insurance claim:
{Overbilling, Underbilling, Non-fraudulent}

● Given a qualitative response and a set of

predictor variables, a classification task needs
to be performed.
5
 Marketing and Psychology
● Branding is a unique psychological trick that creates a lasting
impression in the minds of all of its customers.
● Instance:

● You decide to purchase 1 box of chocolate of a certain brand.

If you have a +ve experience with that after going home, this
will likely make you a customer for life with that one chocolate
brand because your brain will now associate these chocolates
with a positive experience you had.

6
 Discrete Probability Distribution: Binomial
• Use Binomial distribution if two possible (mutually exclusive)
outcomes exist for each case.

• E.g., vote for/against, win/lose, success/failure,

defective/non- defective.

• Binomial Distribution: In n independent “success-failure”

cases with probability of success in each trial being p, the
probability of getting k success out of n trials is given by the
 Characteristics

● The Binomial Distribution is characterized by

n: Number of items/individuals/trials
p: Probability of success

● The mean and variance of Binomial Distribution is given

by
μ = np, = np(1 − p).

● For n=1, , k=0,1.

 Binomial Distribution: Illustrations
● 1: Survey finds 30% of workers take public transportation daily. In a
sample of 10 workers, what is the probability that exactly three
workers take public transportation daily? what is the probability that
at least three workers take public transportation daily?

● 2: A company accepts a lot from a particular supplier if the defective

components in the lot do not exceed 1%. Suppose a random sample
of five items from a recent shipment is tested. If 1% of the shipment
is defective, what is the probability that no items in the sample are
defective? what is the probability that exactly one item in the sample
is defective? What is the expected number of defective items in the
sample? How precise is this expected number of defective items in
the sample?
 Continuous Probability Distribution: Normal

● Normal Distribution is assumed if the data is continuous and

has bell-shaped histogram.

● For e.g.,
● Data on behavioral test scores
● Data on performance scores
● measurement errors.

● Normal Distribution is characterized by mean (μ) and

variance (σ2).
Normal Distribution
 Normal Distribution
● Each choice of (μ,σ) defines a particular normal distribution.
Shape of the normal curve: Symmetric, unimodal, bell-shaped.

● 68-95-99.7% rule, also called Empirical rule states:

Roughly
● 68% of values of X are within 1 SD of mean (μ−σ,μ+σ).
● 95% of values of X are within 2 SD’s of mean (μ−2σ,μ+2σ).
● 99.7% of values of X are within 3 SD’s of mean (μ−3σ,μ+3σ).

● μ − σ and μ + σ are the points of inflection of the curve. Any N(μ,

σ2) distribution with μ = 0 and σ = 1 is called standard normal
distribution.
● For standard normal distribution, N(0,1), the approximate 68-95-
99.7% rule holds.
Empirical Rule
 Normal Distribution: An Illustration
● Suppose packages of cream cheese coming from an
automated processor have weights that are normally
distributed. For one day’s production run, the mean is
8.2 ounces and the standard deviation is 0.1 ounces.
● If the packages of cream cheese are labeled 8
ounces, what proportion of the packages weigh less
than the labeled amount?
● If only 5% of the packages exceed a specified weight
of w, what is the value of w?
● Suppose a package is selected at random from the
day’s production. What is the probability that the
weight of the package is less than 8.3 ounces?
● Suppose 5 packages are selected at random from the
day’s production. What is the probability that at most
one package weighs at least 8.3 ounces?
 Z-Scores
● If x is an observation from a distribution that has mean μ and
standard deviation σ , the standard unit of x is (X−μ)/σ. This value
is commonly referred to as a standard unit (z-score).

● A z-score tells us how many standard deviations above (+) or

below (-) mean an observation is.

● Example: Phyllis takes midterms in two courses. Relative to the

rest of the class, on which test did she do better?

● If X follows N(μ, σ2), then, Z = (X−μ)/σ will follow N(0,1).