How Stan computes the posterior distribution

Bayesian Data Analysis
Bayesian Data Analysis
Public group
Location image of event venue


During this meetup, Stan developer Daniel Lee will take a look under the hood and zoom in on how and why Stan does what it does.

In particular, we will examine how the log posterior, known as lp__ by the insiders, is computed and why the computation is efficient. We'll discuss the practical aspects of only computing things up to a proportionality constant, the importance of the automatic transforms, and how to apply Jacobians for transforms.

Dan is on the Stan core development team, having started with the project at the beginning in 2011. He is also the CTO of Generable where he is building a modern platform for clinical research.

Pizza will be served at 6:30 and we will start at 7.