In this introductory talk, we will discuss how to analyse data using Bayesian inference. In the Bayesian approach, we update our existing beliefs as new data come in instead of testing against a null hypothesis. We will look at how explicitly modelling data generation enables us to derive meaningful insights about latent phenomena. For example, we can infer psychological parameters, decide between non-null alternatives, and work easily with non-Gaussian data and non-linear relationships. To bring these ideas to life without the heavy math, we'll walk through a simple coin toss example alongside a hands-on demonstration using the Python package Bambi. Main Sources:
- McElreath, R. Statistical rethinking: A Bayesian course with examples in R and Stan (CRC press, 2020).
- Kruschke, J. Doing Bayesian data analysis: A tutorial with R, JAGS, and Stan (Academic Press, 2014)
- Abril-Pla O, Andreani V, Carroll C, Dong L, Fonnesbeck CJ, Kochurov M, Kumar R, Lao J, Luhmann CC, Martin OA, Osthege M, Vieira R, Wiecki T, Zinkov R. PyMC: A Modern and Comprehensive Probabilistic Programming Framework in Python (2023)
- Capretto, T. and Piho, C. and Kumar, R. and Westfall, J. and Yarkoni, T. and Martin, O. A. Bambi: A Simple Interface for Fitting Bayesian Linear Models in Python (2022)
Speaker Information: Kai Biermeier, M.Sc. is Lab technician in the TRR 318 "Constructing Explainability". He develops experiments and analyses for different disciplines including: Cognitive Psychology, Psycholiguistics and Human-Computer Interaction. His core interest is the mathematical modelling of cognitive processes especially covert visual attention using Bayesian Inference.
Online, via Zoom
