Prior Predictive Distribution

a prior predictive distribution is denoted as 𝐏(𝑋) where:

  • 𝑋 - data sample

continuous prior predictive distribution

  • 𝐏(𝑋) = ∫𝜃∊𝜣𝐏(𝑋,𝜃)𝑑𝜃
  • 𝐏(𝑋) = ∫𝜃∊𝜣𝐏(𝑋|𝜃)𝐏(𝜃)𝑑𝜃

where:

therefore, when we multiply a prior distribution with the likelihood function and integrate over the range of 𝜃 values, we get the prior predictive distribution

Posterior Predictive Distribution

a posterior predictive distribution is denoted as 𝐏(𝑋ˆ|𝑋) where:

  • 𝑋ˆ - new data sample
  • 𝑋 - observed/given data sample

continuous posterior predictive distribution

  • 𝐏(𝑋ˆ|𝑋) = ∫𝜃∊𝜣𝐏(𝑋ˆ,𝜃|𝑋)𝑑𝜃
  • 𝐏(𝑋ˆ|𝑋) = ∫𝜃∊𝜣𝐏(𝑋ˆ|𝜃,𝑋)𝐏(𝜃|𝑋)𝑑𝜃
  • 𝐏(𝑋ˆ|𝑋) = ∫𝜃∊𝜣𝐏(𝑋ˆ|𝜃)𝐏(𝜃|𝑋)𝑑𝜃 # normally when we condition on 𝜃, 𝑋ˆ is conditionally independent to 𝑋 given 𝜃

where:

therefore, when we multiply a posterior distribution with the likelihood function and integrate over the range of 𝜃 values, we get the posterior predictive distribution

Prior/Posterior Predictive Distributions - Video

Subpages