Model of normally distributed log income:
Model of normally distributed log income:
Model of normally distributed log income:
Pr(μ) prior
Pr(μ | Y) posterior
Pr(σ) prior
Pr(σ | Y) posterior
Pr(μ,σ | Y) posterior
(Marginal) posterior
distribution of μ
Pr(μ | Y)
So many normal distributions!
This can get confusing!
Pr(Y | μ)
Pr(μ)
Pr(μ | Y)
Figures by Peter McMahan (source code)
Image copyright Disney
- Nearly normal in shape - Very narrow bounds - Lumpy because built from posterior