CSSS Seminar: Wed 11/12, Daniel Lee (Columbia Statistics) “Stan: A Platform for Bayesian Inference”

Center for Statistics and the Social Sciences (CSSS) Seminar

Speaker: Daniel Lee (Columbia Statistics)

Title: Stan: A Platform for Bayesian Inference

Time: 12:30 pm on Wednesday, November 12, 2014

Place: Savery 409

Abstract: Stan is a high-level statistical modeling language designed for expressing a wide variety of models. By default, Stan performs full Bayesian inference using Markov chain Monte Carlo (MCMC). There are interfaces to many popular computing environments including R, Python, command line, Matlab, and Julia. Stan’s default MCMC algorithm is an adaptive version of Hamiltonian Monte Carlo (HMC). HMC requires not only
the computation of the (log) joint probability function, but also the gradients with respect to all parameters. This talk will describe the modeling language, provide some intuition for HMC and some of the implementation challenges, and a provide a couple examples using Stan. Stan is an open-source software project and relies on the contributions of many individuals, including Andrew Gelman, Bob Carpenter, Matt Hoffman, Daniel Lee, Ben Goodrich, Michael Betancourt, Marcus Brubaker, Jiqiang Guo, Peter Li, Allen Riddell, Marco Inacio, Jeffrey Arnold, and Mitzi Morris.

This entry was posted in Campus Event, Seminars. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s