Processing math: 100%

Statistics - Gamma Distribution


Advertisements

The gamma distribution represents continuous probability distributions of two-parameter family. Gamma distributions are devised with generally three kind of parameter combinations.

  • A shape parameter k and a scale parameter θ.

  • A shape parameter α=k and an inverse scale parameter β=1θ, called as rate parameter.

  • A shape parameter k and a mean parameter μ=kβ.

Gamma Distribution

Each parameter is a positive real numbers. The gamma distribution is the maximum entropy probability distribution driven by following criteria.

Formula

E[X]=kθ=αβ>0 and is fixed.E[ln(X)]=ψ(k)+ln(θ)=ψ(α)ln(β) and is fixed.

Where −

  • X = Random variable.

  • ψ = digamma function.

Characterization using shape α and rate β

Probability density function

Probability density function of Gamma distribution is given as:

Formula

f(x;α,β)=βαxα1exβΓ(α) where x0 and α,β>0

Where −

  • α = location parameter.

  • β = scale parameter.

  • x = random variable.

Cumulative distribution function

Cumulative distribution function of Gamma distribution is given as:

Formula

F(x;α,β)=x0f(u;α,β)du=γ(α,βx)Γ(α)

Where −

  • α = location parameter.

  • β = scale parameter.

  • x = random variable.

  • γ(α,βx) = lower incomplete gamma function.

Characterization using shape k and scale θ

Probability density function

Probability density function of Gamma distribution is given as:

Formula

f(x;k,θ)=xk1exθθkΓ(k) where x>0 and k,θ>0

Where −

  • k = shape parameter.

  • θ = scale parameter.

  • x = random variable.

  • Γ(k) = gamma function evaluated at k.

Cumulative distribution function

Cumulative distribution function of Gamma distribution is given as:

Formula

F(x;k,θ)=x0f(u;k,θ)du=γ(k,xθ)Γ(k)

Where −

  • k = shape parameter.

  • θ = scale parameter.

  • x = random variable.

  • γ(k,xθ) = lower incomplete gamma function.