Linking data to parameters
Applying Bayesian statistics to real-world problems
Welcome to the third workshop of the BayesCog course!
In this workshop, we’ll build on our understanding of probability and Bayes’ theorem to explore how we can construct and implement Bayesian models. We’ll start with a simple but powerful example - the binomial model - and use it to demonstrate key concepts in Bayesian modeling. We will explore the computational burden of Bayesian statistics - calculating posterior distributions - first through grid approximation, before exploring Markov chain Monte Carlo (MCMC) methods - powerful computational techniques which allow us to sample from complex posterior distributions that would be impossible to compute directly.
Topics for this workshop include:
- Understanding how data and parameters are linked in Bayesian inference
- Learning about likelihood functions and how to choose them
- Exploring the binomial model through a practical example
- Using grid approximation to estimate posterior distributions
- An introduction to Markov chain Monte Carlo
Model code and R scripts for this workshop are located in the (/workshops/02.binomial_globe) directory. Remember to use the R.proj file within each folder to avoid manually setting directories!
The copy of this workshop notes can be found on the course GitHub page.