Join us for the first RSS Edinburgh Local Group event of 2025, joint with the R-Edinburgh community. In this event we will welcome two speakers, Isabella Deutsch and Jordan Richards, who will discuss their favourite ways of working with Bayesian Statistics in the R programming language.
Isabella Deutsch: R You Ready for ABC? Likelihood-Free Bayesian Inference in Action.
Isabella Deutsch recently completed a PhD in Statistics at the University of Edinburgh and now works as a Data Scientist at Smart Data Foundry. She is passionate about being a translator between people and data, such as at her show at the Fringe 2024.
Approximate Bayesian Computation (ABC) offers an intuitive -- and likelihood-free -- way to talk about Bayesian Statistics. We investigate the use of ABC in the context of point processes with missing data, demonstrating how it can approximate posterior distributions using simulation alone. Drawing from real-world applications, we highlight the strengths and limitations of ABC in practice. Finally, we explore some code snippets that show how easy it is to get started with ABC in R.
Jordan Richards: Fast and Amortised Bayesian Inference with NeuralEstimators
Jordan Richards is a lecturer of statistics in the School of Mathematics at The University of Edinburgh. His main research interest is extreme value analysis, and its intersection with spatial statistics and machine learning.
Neural estimators are neural networks that transform data into parameter point estimates. These estimators are likelihood-free and amortised, in the sense that, after an initial setup cost, inference from observed data can be made in a fraction of the time required by conventional approaches, e.g., MCMC or maximum likelihood estimation. They are also approximate Bayes estimators and, therefore, are often referred to as neural Bayes estimators. We present the user-friendly R package NeuralEstimators, which interfaces with the Julia package NeuralEstimators.jl, and allows for the development and application of neural Bayes estimators. This package caters for any model for which simulation is feasible by allowing the user to implicitly define their model via simulated data. No likelihood or long MCMC chains required!
Book now