Across multiple disciplines, computer models have become vitals tools for research. Computer models enable researchers to run simulation-based experiments, to make predictions under different scenarios and to assess the impact of policies. A common thread amongst the various modelling approaches is the presence of uncertainty. The quantification and reduction of uncertainty can result in more accurate predictions and, thus, more robust decision-making.

This webinar will provide an introduction to the Bayesian calibration of computer models using R. The session will include:

- A brief description of Bayes' Theorem and Bayesian calibration theory

- An introduction to the powerful R package rstan and its basic functionality

- A step-by-step demonstration of Bayesian calibration using a simple model

The code and data will be shared with the attendees before the event. Attendees should make sure to have R and rstan installed before the event. (Please note that rstan is not currently supported on R version 4.2, but you could install the previous R version 4.1.3 to use for this webinar, even if you already have the newest version installed. Please see here for more detailed information on this issue and here for a detailed guide on installing rstan.)