Jonah Gabry is a Statistician at Columbia University, collaborating with Andrew Gelman on methods and software for Bayesian data analysis. He is a member of the Stan core development team. The Stan project develops free and open-source software for Bayesian statistical modeling that interfaces with the most common data analysis platforms (R, Python, MATLAB, Stata, etc.). The Stan language and inference algorithms are used throughout academia and industry for everything from clinical drug trials, to professional sports analytics, to gravitational wave detection. He is co-author of the rstan and rstanarm R packages, which provide interfaces to Stan. He is author and maintains the shinystan and bayesplot packages for model visualization, and the loo R package for model comparison. In addition to developing statistical software, Jonah is affiliated with Columbia’s Institute of Social and Economic Policy and Research, and the Columbia Population Research Center, where he is an advisor on statistical issues related to the collection and analysis of survey data. In this talk, Jonah Gabry will present an introduction to Stan, and discuss how Stan’s algorithms leverage connections between high-dimensional probability distributions and physics to make it easier for researchers to estimate the increasingly complex models of interest in the social, biological, and physical sciences.
Title: Stan: A Software Ecosystem for Modern Bayesian Inference