Posterior Distribution

Gist

Used for Bayesian analysis

Definition

Using the Bayes formula we know that:

which for our observed data $y$ and model parameter $\theta$ can be rewritten as:

However, analytical expressions are impossible a lot of times so we need to be able to simulate from the posterior distribution through methods like Markov Chain Monte Carlo.