- Location
- HS790
- Series/Type
- Alumni Event, DLSPH Event, Faculty/Staff Event, Student Event
- Format
- In-Person
- Dates
- March 25, 2024 from 1:00pm to 2:00pm
Links
Presented by Rose Garrett.
Bayesian methods produce clear and easily interpretable results, allow us to incorporate all the available information through the use of priors, and offer a flexible template for setting up complex models. In this workshop, we will explore some practical aspects of Bayesian modelling using two popular probabilistic programming languages: JAGS and Stan.
The session will begin with a quick overview of how to set up models in JAGS and Stan. We will then explore the process of coding custom densities for the likelihood in both languages.
We will also explore a general framework for setting up different types of joint models, including the traditional joint longitudinal and time-to-event models.
We will also discuss quirks of working with the two different languages, and other practical tips for Bayesian modelling.