Types of Distribution

Normal Distribution
The bell curve is another name for the normal distribution. It occurs naturally in a number of situations;
for example, the normal distribution can be seen in GRE and SAT experiments. There are also a variety of
classes that adopt the normal distribution pattern. As a consequence, it is frequently used in statistics,
company, and government agencies such as the FDA:

 Measurement errors.
 Points on a test.
 Salaries.
 Height of people.
 Blood pressure.
 IQ scores.
Properties of a normal distribution
 The curve remains symmetric at the center.
 The area under the curve is 1.
 The mean, median, and mode are always equal.
 Exactly half value is on the left of the center and other on the right.
T- Distribution
In statistics, it is one of the most critical distributions. The probability distribution is also known as
Student's t- distribution. When the sample size is small, this method is used to estimate the population's
parameters. Often unknown is the population's standard deviation.

Properties of t-distribution
 Like normal distribution, the student distribution has bell-shaped and symmetric with
zero mean.
 The range of student distribution from – ∞ to ∞ (infinity).
 The t distribution’s shape changes with the degree of freedom.
 The variance is always more than one, and it can be represented when the degree of
freedom V>=3 and given: Var (t) = [v/v -2].
 It is not packed that much at the center but higher at trials; therefore, its shape is like
platykurtic.
 The dispersion of t distribution is much more than the normal distribution. As the size of
the sample ‘n’ increases, it is considered as a normal distribution. Here, the given sample
size is taken larger than n>=30.

Uniform distribution
The uniform distribution is the most basic type of continuous distribution. It has a rectangular
distribution with a constant likelihood. It also means that each value has the same distribution
length. Which has an equal likelihood of occurring. This function, on the other hand, belongs to
the group of maximum entropy probability distributions.
Characteristics of uniform distribution
 The density function combines to unity.
 Each of its input function has equal weightage.
 The uniform function’s mean is given by:

 The variance of the uniform distribution is given by:

Bernoulli distribution
A Bernoulli distribution is a discrete probability distribution that describes a random trial with two
outcomes. A single coin flip, for example, is a special case with the value n = 1.

Characteristics of Bernoulli distribution

 The number of trials that have to be performed under a single experiment must be
predefined.
 Each trail must have two outcomes that are success or failure.
 The success’ probability in each experiment should be the same.
 The experiment should be independent of each other, which means the outcome of one
trial is not affected by the other’s trial outcome.
Properties 

The expected value of the randomly selected variable is given by E(x) = p and can be
derived:

E(x) = 0*(1-p) + 1 * p = p
The variance of the Bernoulli variable is given by p*(1-p) and is given as:
Var(X) = p – p2 = p*(1-p)
Binomial distribution
Under a set of assumptions or parameters, it is a probability distribution that concludes
a value that takes one of two independent values. Furthermore, the predictions of the
binomial distribution must yield a single outcome with the same probability of success.
And each trail must be distinct from the others.
Properties of a binomial distribution
 When an experiment has independent trails, and each of them has two results that are
success and failure.
 The binomial distribution is also called as bi-parametric distribution. As it is classified by
two parameters n and p.
 The mean value of this is:
μ  = np

 The binomial distribution’s variance is given by:

σ²  = npq

 The value of p and q is always less than or equal to 1, or we can say that the
variance must be less than its mean value.

npq < np

Poisson distribution
It is a method that is used to estimate a particular likelihood of an occurrence when the value of
the event is known. The Poisson distribution informs us how often a given number of events will
occur in a given amount of time.

Properties of Poisson distribution

 The expected value and variance of the random variable are equivalents to λ.
 The absolute deviation is associated with the mean is given by:

 The coefficient of variance is , whereas the index dispersion is .

 The expected value of the Poisson distribution is decomposed by underlying the product
of intensity and exposure.
 The mean of the Poisson distribution is given by “m”.

Exponential distribution
The time between the trails in a Poisson process is defined by a negative exponential
distribution, also known as a negative exponential distribution. The exponential distribution and
the Poisson distribution have a relationship.
Some of the formulas of it
 An exponential random variable’s expected value is given by:

 An exponential random variable’s variance value is given by:

 The exponential random variable’s moment generating function is given by:

  An exponential random variable’s characteristic function is given by:

Beta distribution
It's a group of continuous probability distributions that fall between [0,1] and are
represented by the letters alpha and beta. This model is often used for the model that
has an ambiguity in the likelihood of a random experiment succeeding. It also provides
a powerful tool that uses simple statistics to measure the degree of confidence in the
completion time.
Properties of beta distribution

There are a few properties that can satisfy these distributions are:

The terms to measure the central tendency are:

 Mean
 Harmonic Mean
 Mode
 Median
 Geometric Mean
The terms to measure the statistics dispersion are:
 Geometric variance and covariance
 Variance
 Mean absolute deviation around the mean
 Mean absolute difference
Beta-binomial distribution
It's the most common Bayesian model, and it's commonly used in fields including
intelligence research, epidemiology, and marketing. If the likelihood of success is p and
the form of the beat binomial parameter is α > 0 and β > 0, the distribution is said to be
beta-binomial.
The parametric shape can be defined as the success probability:
 A distribution can approach a binomial distribution for the larger value of α and β.
 The value of discrete uniform distribution equals the distribution from 0 to n, if the value
of both α and β is equal to 1.
 For the value of n = 1, the beta-binomial distribution is the same value as that of
Bernoulli distribution.
The key difference between a beta-distribution and a binomial distribution is that in a
binomial distribution, p is always constant for a series of trials, while in a beta-binomial
distribution, p is not fixed and switches trail to trail.
Log-normal distribution
f the log to the power is normally distributed, then the variable is taken as lognormally
distributed. Or we can say that ln(x) is normally distributed and that the variable x is
assumed to have a log-normal distribution.
Properties of the log-normal distribution
 The expected value or the mean of distribution offers useful data about what an average
would expect from a repeated trail number.
 The median of a log-normal distribution is another consideration of central tendency,
and it is useful for outliers that help the means to lead.
 The distribution’s mode is a value with the highest probability of occurs.
 How to spread out the information can be measured by the variance. The square root of
the variance and the standard deviation are useful as these have the same unit of the
data.
These values are much easier to measure for a continuous probability distribution. But as its
measure includes a fair amount of calculus, the description can be brief.