Suppose that x = {x1, . . . , xm} is an independant and identically distributed sample from N(θ1, σ2). y = {y1, . . . , yn} is an independant and identically distributed sample from N(θ2, σ2). x and y are independent. The unknown parameters are θ1∈ (−∞, ∞), θ2∈ (−∞, ∞), and σ2∈ (0, ∞).
a) Under the improper prior p(θ1, θ2, σ2) ∝ (σ2)−2, find the joint posterior distribution of (θ1, θ2, σ2) by finding the following three distributions:
1. p(θ1|σ2, x, y),
2. p(θ2|σ2, x, y),
3. p(σ2|x, y),
and show that p(θ1, θ2, σ2|x, y) ∝ p(θ1|σ2, x, y) × p(θ2|σ2, x, y) × p(σ2|x, y).
b) Following Part (a), let δ = θ1− θ2. Find a suitable change of variable for δ, such that the marginal posterior of the transformed variable from δ has a Student’s t distribution.
Get Answers For Free
Most questions answered within 1 hours.