ABSTRACT
Generic description: Number of Bernoulli trials until the first success Range of values: x = 1, 2, 3, . . . Parameter: p ∈ (0, 1), probability of success Probability mass function: P (x) = p(1− p)x−1 Cdf: 1− (1 − p)x
Expectation: μ = 1 p
Variance: σ2 = 1− p p2
Relations: Special case of Negative Binomial(k, p) when k = 1 A sum of n independent Geometric(p) variables is
Negative Binomial(n, p)
Negative Binomial(k, p)
Generic description: Number of Bernoulli trials until the k-th success Range of values: x = k, k + 1, k + 2, . . . Parameters: k = 1, 2, 3, . . ., number of successes
p ∈ (0, 1), probability of success Probability mass function: P (x) =
( x− 1 k − 1
) (1 − p)x−kpk
Expectation: μ = k
p
Variance: σ2 = k(1− p)
p2
Relations: Negative Binomial(k, p) is a sum of n independent Geometric(p) variables
Negative Binomial(1, p) = Geometric(p)
Poisson(λ)
Generic description: Number of “rare events” during a fixed time interval Range of values: x = 0, 1, 2, . . . Parameter: λ ∈ (0,∞), frequency of “rare events”
Probability mass function: P (x) = e−λ λx
x! Cdf: Table A3 Expectation: μ = λ Variance: σ2 = λ Relations: Limiting case of Binomial(n, p) when
n →∞, p → 0, np → λ Table: Appendix, Table A3
Beta(α, β)
Generic description: In a sample from Standard Uniform distribution, it is the distribution of the kth smallest observation
Range of values: 0 < x < 1 Parameter: α, β ∈ (0,∞), frequency of events, inverse scale parameter
Density: f(x) = Γ(α + β) Γ(α)Γ(β)
xα−1(1 − x)β−1
Expectation: μ = α/α + β Variance: σ2 = αβ/(α + β)2(α + β + 1)
Relations: As a prior distribution, Beta family is conjugate to the Binomial model.