We will be talking about Stan at the Seattle useR Meetup on Tuesday, May 17 and at the Open Data Science Boston conference on Friday, May 20 and Sunday May 22.
The Seattle useR meetup organized by Zach Stednick is filling up so if you are in the Seattle area next week, please stop by. Here is the abstract:
We will build up a simple Stan model from scratch to demonstrate various parts of the Stan program and will also present a multi-level, overdispersed Poison model that can be used for pricing products in a retail setting. We fit the latter model with the rstanarm package, which can be thought of as a fully Bayesian equivalent to lme4.
Explicit and implicit models are widely used for analysis and prediction in data science. Stan is a multi-language estimation tool ideal for applications where models matter. Stan provides a simple language for specifying and combining model building blocks such as random effects, hierarchical structures, differential equations and mixture models. These building blocks are contained in a probabilistic programming language that allows users to specify (nearly) arbitrary statistical models and backed by state-of-the-art optimization and sampling algorithms efficient enough for Big Model analyses. Stan gives data scientists the freedom to apply a variety of competing models and evaluate their performance with their favorite munging and analysis tools. In this workshop participants will be introduced to the Stan language and led through the model building and criticism cycle on a real-world dataset.
At the same conference, Eric Novik will be giving another introductory Stan talk on Sunday, May 22.
We are always eager to meet new and existing Stan users, so please come by and say hello.