3 Common Families of Distributions

3.1 Introduction

Statistical distributions are used to model populations; as such, we usualy deal with a family of distributions rather than a single distribution.

This family is indexed by one or more parameters, which allow us to vary certian characteristics of the distribution while staying with one functional form.

3.2 Discrete Distributions

A random variable X is said to have a discrete distribution if the range of X, the sample space, is countable.

Discrete Uniform Distribution

A random variable X has a discrete uniform (1,N) distribution if

\[P(X=x|N) = \frac{1}{N}, x = 1,2,3,...,N\],

where N is a specified integer.

\[E(X) = \sum_{x=1}^N x \frac{1}{N} = \frac{N+1}{2} \]

\[Var(X) = E(X^2) - [E(X)]^2 = \frac{(N+1)(2N+1)}{6} - (\frac{N+1}{2})^2 = \frac{(N+1)(N-1)}{12}\]

Hypergeometric Distribution

Binomial Distribution

Poisson Distribution

Negative Binomial Distribution

Geometric Distribution

3.3 Continuous Distributions

Uniform Distribution

Gamma Distribution

Normal Distribution

Beta Distribution

Cauchy Distribution

Lognormal Distribution

Double Exponential Distribution

3.4 Exponential Families

A family of pdfs or pmfs is called an exponential family if it can be expressed as

\[f(x|\boldsymbol{\theta}) = h(x)c(\boldsymbol{\theta})exp(\sum_{i=1}^kw_i(\boldsymbol{\theta})t_i(x))\].

Here \(h(x) \geq 0\) and \(t_1(x),...,t_k(x)\) are real-valued functions of the observation x (they cannot depend on \(\boldsymbol{\theta}\)), and \(c(\boldsymbol{\theta}) \geq 0\) and \(w_1(\boldsymbol{\theta}), ..., w_k(\boldsymbol{\theta})\) are real-valued functions of the possible vector-valued parameter \(\boldsymbol{\theta}\) (they cannot depend on x).

Many common families are exponential families, including the continuous families (normal, gamma, and beta), and the discrete families (binomial, Poisson, and negative binomial).

3.5 Location and Scale Families

Three techniques for constructing families of distributions. The three types of families are called location families, scale families, and location-scale families.

3.6 Inequalities and Identities

3.6.1 Probability Inequalities

The most famous, and perhaps most useful, probability inequality is Chebychev’s Inequality.

3.6.2 Identities

3.7 Miscellanea