Bayesian inference with RStan
Details
LINK TO PRE WORKSHOP INSTRUCTIONS/WHERE TO GO HERE
https://docs.google.com/document/d/1DlFf8TJq-BqYxUmWf2G9LGBHOHJOU4zsTUWCkI4m3hE/edit?usp=sharing
STOP THE PRESS!
Michael Betancourt is visiting Sydney and has offered to run a Stan workshop for us… eeeeep!! We couldn’t turn down this opportunity.
How do we learn effectively from data? As scientists, engineers and analysts, this is a constant refrain. One answer to this question has traditionally been "just use big data sets", yet this is not always practical nor even relevant. In real life, our statistical inferences are restricted not because we have too little data but because we ignore the systematic structure of our data. Only by carefully modelling this structure can we take fully advantage of the data - big or small - available to us.
In this course we provide an introduction to the basics of Bayesian inference, and the Stan software used to implement Bayesian methods. The power of this approach lies in the flexibility of Stan - it provides a an expressive modelling language for specifying customised, bespoke models and implementing state-of-the-art algorithms to draw subsequent Bayesian inferences. As an R-Ladies event, the workshop runs through interactive exercises run with RStan, the R interface to Stan.
Prerequisites
This event is best geared for intermediate/advanced R users. We assume you are interested in Bayesian statistics and are familiar with the basics of calculus and linear algebra. Conceptual understanding of probability theory and conditional probability will be useful; check out content on Michael Betancourt's website.
https://betanalpha.github.io/assets/case_studies/probability_theory.html
https://betanalpha.github.io/assets/case_studies/conditional_probability_theory.html
In order to participate in the interactive exercises attendees must bring a laptop with R and the latest version of RStan installed (https://cran.r-project.org/web/packages/rstan/index.html).
Please verify that you can run the 8schools model as discussed in the RStan Quick Start Guide (https://github.com/stan-dev/rstan/wiki/RStan-Getting-Started) and report any installation issues on the Stan Forums (https://discourse.mc-stan.org) as early as possible.
For beginners who need help with the above, there will be a pre-Stan Hacky Hour the week before the event.
Bring your lunch on Sunday, we will kick off around 12:30 and aim to be done by 5:30pm.