7.1 The big idea 1/2
We are going to use a Normal-Normal model:
Y|μ∼Norm(μ,0.752)
μ∼Norm(0,12)
Observed outcome 6.25:
μ|(Y=6.25)∼Norm(4,0.62)
Main idea: chain need to spend more time around μ value. Remember μi+1 is dependant of μi.
How are we going to visit every part of the posterior dustribution:
step 1 : propose a random location μ′ (I prefer μproposal) for the nex stop
step 2 : Decide whether to:
- go to the proposed location: μproposal=μi+1
- stay at the current location: μ=μi+1
Monte Carlo algorithm:
step 1 propose location: draw μ from posterior model with pdff(μ|y)
step 2 : Go there