An overview of Distributions

A distribution shows the possible values a random variable can take and how frequently they occur.

Important Notation for Distributions:

Y actual outcome Population Sample

Mean 𝜇 𝑥ҧ
y one of the possible outcomes Variance 𝜎2 𝑠2
𝑃(𝑌 = 𝑦) is equivalent to p(𝑦). Standard 𝜎 𝑠
Deviation

We call a function that assigns a

probability to each distinct outcome
in the sample space, a probability
function.

𝜇−𝜎 𝜇 𝜇+𝜎
 Types of Distributions

Certain distributions share characteristics, so we separate them into types. The well-defined types of
distributions we often deal with have elegant statistics. We distinguish between two big types of
distributions based on the type of the possible values for the variable – discrete and continuous.

Discrete Continuous

• Have a finite number of outcomes. • Have infinitely many consecutive possible

• Use formulas we already talked about. values.
• Can add up individual values to determine • Use new formulas for attaining the probability
probability of an interval. of specific values and intervals.
• Can be expressed with a table, graph or a • Cannot add up the individual values that make
piece-wise function. up an interval because there are infinitely
• Expected Values might be unattainable. many of them.
• Graph consists of bars lined up one after the • Can be expressed with a graph or a
other. continuous function.
• Graph consists of a smooth curve.
 Discrete Distributions

Discrete Distributions have finitely many different possible outcomes. They possess several key
characteristics which separate them from continuous ones.

Key characteristics of discrete distribution

• Have a finite number of outcomes.
• Use formulas we already talked about.
• Can add up individual values to determine
probability of an interval.
• Can be expressed with a table, graph or a piece-wise
function.
• Expected Values might be unattainable.
• Graph consists of bars lined up one after the other.
• 𝑷 𝒀≤𝒚 =𝑷 𝒀<𝒚+𝟏

Examples of Discrete Distributions:

• Discrete Uniform Distribution
• Bernoulli Distribution
• Binomial Distribution
• Poisson Distribution
 Uniform Distribution

A distribution where all the outcomes are equally likely is called a Uniform Distribution.

Notation:
• 𝒀~ 𝑼(𝒂, 𝒃)
• * alternatively, if the values are categorical, we simply
indicate the number of categories, like so: 𝒀~ 𝑼(𝒂)

Key characteristics
• All outcomes are equally likely.
• All the bars on the graph are equally tall.
• The expected value and variance have no predictive
power.

Example and uses:

• Outcomes of rolling a single die.
• Often used in shuffling algorithms due to its fairness.
 Bernoulli Distribution

A distribution consisting of a single trial and only two possible outcomes – success or failure is called a
Bernoulli Distribution.

Notation:
• 𝒀~ 𝑩𝒆𝒓𝒏(𝒑)

Key characteristics
• One trial.
• Two possible outcomes.
• 𝑬 𝒀 =𝒑
• 𝑽𝒂𝒓 𝒀 = 𝒑 × (𝟏 − 𝒑)

Example and uses:

• Guessing a single True/False question.
• Often used in when trying to determine what we
expect to get out a single trial of an experiment.
 Binomial Distribution

A sequence of identical Bernoulli events is called Binomial and follows a Binomial Distribution.

Notation:
• 𝒀~ 𝑩(𝒏, 𝒑)

Key characteristics
• Measures the frequency of occurrence of one of the
possible outcomes over the n trials.
• 𝑷 𝒀 = 𝒚 = 𝑪 𝒚, 𝒏 × 𝒑𝒚 × 𝟏 − 𝒑 𝒏−𝒚
• 𝑬 𝒀 =𝒏×𝒑
• 𝑽𝒂𝒓 𝒀 = 𝒏 × 𝒑 × (𝟏 − 𝒑)

Example and uses:

• Determining how many times we expect to get a
heads if we flip a coin 10 times.
• Often used when trying to predict how likely an
event is to occur over a series of trials.
 Poisson Distribution

When we want to know the likelihood of a certain event occurring over a given interval of time or distance
we use a Poisson Distribution.

Notation:
• 𝒀~ 𝑷𝒐(λ)

Key characteristics
• Measures the frequency over an interval of time or
distance. (Only non-negative values.)
𝒚 −λ
λ 𝒆
• 𝑷 𝒀=𝒚 =
y!
• 𝑬 𝒀 =λ
• 𝑽𝒂𝒓 𝒀 = λ

Example and uses:

• Used to determine how likely a specific outcome is,
knowing how often the event usually occurs.
• Often incorporated in marketing analysis to
determine whether above average visits are out of
the ordinary or not.
 Continuous Distributions

If the possible values a random variable can take are a sequence of infinitely many consecutive values, we
are dealing with a continuous distribution.

Key characteristics
• Have infinitely many consecutive possible values.
• Cannot add up the individual values that make up
an interval because there are infinitely many of them.
• Can be expressed with a graph or a continuous
function. Cannot use a table, be
• Graph consists of a smooth curve.
• To calculate the likelihood of an interval, we need
integrals.
• They have important CDFs.
• 𝑷 𝒀 = 𝒚 = 0 for any individual value y.
• 𝑷 𝒀<𝒚 =𝑷 𝒀≤𝒚
 Normal Distribution

A Normal Distribution represents a distribution that most natural events follow.

Notation:
• 𝒀~ 𝑵(μ, σ𝟐)

Key characteristics
• Its graph is bell-shaped curve, symmetric and has
thin tails.
• 𝑬 𝒀 =μ
• 𝑽𝒂𝒓 𝒀 = σ𝟐
• 68% of all its values should fall in the interval:
• (μ − 𝝈, 𝝁 + 𝝈)

Example and uses:

• Often observed in the size of animals in the
wilderness.
• Could be standardized to use the Z-table.
 Standardizing a Normal Distribution

To standardize any normal distribution we need to transform it so that the mean is 0 and the variance and
standard deviation are 1.

Ensures mean is 0.
Using a 𝑦−𝜇
transformation to z=
create a new 𝜎
random variable z.
Ensures standard
deviation is 1.

Importance of the Standard Normal Distribution.

• The new variable z, represents how many standard deviations away from the mean, each corresponding
value is.
• We can transform any Normal Distribution into a Standard Normal Distribution using the
transformation shown above.
• Convenient to use because of a table of known values for its CDF, called the Z-score table, or simply the
Z-table.
 Students’ T Distribution

A Normal Distribution represents a small sample size approximation of a Normal Distribution.

Notation:
• 𝒀~ 𝒕 (𝒌)

Key characteristics
• A small sample size approximation of a Normal
Distribution.
• Its graph is bell-shaped curve, symmetric, but has fat
tails.
• Accounts for extreme values better than the Normal
Distribution.
𝒌
• If k>1: 𝑬 𝒀 = μ and 𝑽𝒂𝒓 𝒀 = 𝒔𝟐 ×
𝒌−𝟐
Example and uses:
• Often used in analysis when examining a small
sample of data that usually follows a Normal
Distribution.
 Chi-Squared Distribution

A Chi-Squared distribution is often used.

Notation:
• 𝒀~ 𝝌𝟐(𝒌)

Key characteristics
• Its graph is asymmetric and skewed to the right.
• 𝑬 𝒀 =𝒌
• 𝑽𝒂𝒓 𝒀 = 𝟐𝒌
• The Chi-Squared distribution is the square of the t-
distribution.

Example and uses:

• Often used to test goodness of fit.
• Contains a table of known values for its CDF called
the 𝝌𝟐-table. The only difference is the table shows
what part of the table
 Exponential Distribution

The Exponential Distribution is usually observed in events which significantly change early on.

Notation:
PDF • 𝒀~ 𝑬𝒙𝒑 (𝝀)

Key characteristics
• Both the PDF and the CDF plateau after a certain
point.
𝟏
• 𝑬𝒀 =
𝝀
𝟏
• 𝑽𝒂𝒓 𝒀 = 𝝀𝟐
• We often use the natural logarithm to transform the
values of such distributions since we do not have a
table of known values like the Normal or Chi-
Squared.
CDF
Example and uses:
• Often used with dynamically changing variables, like
online website traffic or radioactive decay.
 Logistic Distribution

The Continuous Logistic Distribution is observed when trying to determine how continuous variable inputs
can affect the probability of a binary outcome.

Notation:
• 𝒀~ 𝑳𝒐𝒈𝒊𝒔𝒕𝒊𝒄 (𝝁, 𝒔)
PDF
Key characteristics.
• 𝑬 𝒀 =𝝁 𝟐 𝟐
𝒔 ×𝝅
• 𝑽𝒂𝒓 𝒀 =
𝟑
• The CDF picks up when we reach values near the
mean.
• The smaller the scale parameter, the quicker it
reaches values close to 1.

Example and uses:

CDF • Often used in sports to anticipate how a player’s or
team’s performance can determine the outcome of
the match.
 QUIZ #1

1. What is a distribution?
a) A collection of all the possible values a variable can take and how frequently they occur in the sample space
b) A function which assigns a probability to each value a variable can take
c) The average value of the elements in the data set
d) The likelihood of an event occurring

2. What is the difference between sample data and population data?

a) Sample data represents the entire data we have, while population data is only some part of it
b) The terms sample data and population data re interchangeable and mean a part of the data
c) The terms sample data and population data are interchangeable and mean the entire data set
d) Sample data represents some part of the data while population data is the same as all the data.

3. Which of the following is expressed with the letter s?

a) Population mean
b) Population variance
c) Population standard deviation
d) Sample mean
e) Sample standard deviation
 QUIZ #2

11. What are the two main types of distributions based on the type of data we have?
a) Distinct and Continuous
b) Discrete and Continuous
c) Finite and Infinite
d) Discrete and Infinite

2. Why do we group distributions into types?

a) We don’t group distributions into types
b) Because they share certain features
c) Because they have different numbers of possible values
d) None of the above
 QUIZ #3

1. How do we compute the probability of an interval for a discrete distribution?

a) We can’t
b) We add up the values of the individual outcomes in the interval
c) We add up the probabilities of each individual outcome in the interval occurring
d) We express the probability distributin in a table and add up the probabilities of the two end points of the
interval

2. Which of the following is true for any variable Y following a discret distribution?
a) P(Y = y) = P(Y < y)
b) P(Y = y) > P(Y <= y)
c) P(Y < y) = P(Y <= y+1)
d) P(Y <+ y) = P(Y < y+1)
 QUIZ #4

1. What does it mean for all possible values form a Uniform Distribution to be “equiprobable”?
a) All the values have the same probability of occurring
b) Values equally away from the mean are equally likely to occur
c) All the possible outcomes are equal to the mean
d) None of the above

2. Why do the expected value and variance have no predictive power for a Discrete Uniform Distribution?
a) They do hold predictive power
b) Because the expected value will posses the same probability as any of the other values in the sample space
so we can’t create prediction intervals
c) Because the standard deviation is not an integer value, so we cannot construct a prediction interval
centered around the mean
d) None of the above
 QUIZ #5

1. What are the two key characteristics of the Bernoulli Distribution?

a) A single trial and only tow possible outcomes
b) Several trials and only two possible outcomes
c) Several trials and several possible outcomes
d) A single trial and several possible outcomes
 QUIZ #6

1. What is the relationship between Binomial and Bernoulli events?

a) A Binomial event is a sequence of identical Bernoulli events
b) A Bernoulli event is a sequence of identical Binomial events
c) There is no relationship between the two types of events
d) None of the above
 QUIZ #7

1. Why does the graph of the Poisson Distribution start from 0 on the X-axis?
a) Because no outcome can have a probability of occurring lower than 0
b) Because we are measuring the frequency of occurrence over a given interval of time or distance
c) The graph of the Poisson distribution does not start from 0 on the X-axis
d) All of the above
 QUIZ #8

1. Graphically, what is the relationship between the PDF and the CDF of a continuous distribution?
a) The PDF represents the area under the curve of the CDF
b) The CDF represents the area under the curve of the PDFA
c) There is no graphical representation of the relationship between the two
d) The CDF and the PDF are interchangeable definitions, so they express the same curve and area on the
graph
 QUIZ #9

1. Which of the following is a characteristic of the graph of a Normal Distribution?

a) Bell-shaped
b) Symetric
c) Thin tails
d) All of the above
 QUIZ #10

1. What are the two characteristics of a Standard Normal Distribution?

a) Mean of 0 and a variance of 1
b) Mean of 1 and variance of 0
c) Variance of 0 and mean of 1
d) Mean of 0 and variance of -1
 QUIZ #11

1. When do we use a Student’s T Distribution?

a) When we have limited data about a variable, which historically follows a Normal Distribution
b) When we have limited data about a variable, regardless of other features of the data set
c) When the variable possesses characteristics of a Normal Distribution and we have an excess of 30
observations
d) None of the above
 QUIZ #12

1. What is one of the features of the graph of a Chi-Squared Distribution?

a) Bell-shaped
b) Symmetric
c) Thin tails
d) None of the above
 QUIZ #13

1. Which of the following is correct about the graphs of the Exponential distribution?
a) The CDF plateaus near 1 mark and the PDF plateaus around the 0 mark
b) The PDF plateaus near 1 mark and the CDF plateaus around the 0 mark
c) The curve of the CDF resembles a boomerang with its handles lining up with the X- and the Y-axes
d) None of the above
 QUIZ #14

1. What is true about the CDF of a Logistic distribution?

a) It follows an S-shape and plateaus around the 1 value
b) The probability drastically starts to increase once we reach values close to the mean
c) The steeper the curve is, the faster it reaches value close to absolute certainty (1)
d) All of the above