Dogucu, Mine; Johnson, Alicia; (2022) Supporting Bayesian Modeling With Visualizations. In: Proceedings of the 11th International Conference on Teaching Statistics (ICOTS11 2022). International Association for Statistical Education: Rosario, Argentina.

Preview

Text ICOTS11_222_DOGUCU.pdf - Published Version Download (368kB) | Preview

Abstract

With computational advances, Bayesian modeling is becoming more accessible. But because Bayesian thinking often differs from learners’ previous statistics training, it can be challenging for novice Bayesian learners to conceptualize and interpret the three major components of a Bayesian analysis: the prior, likelihood, and posterior. To this end, we developed an R package, bayesrules, which provides tools for exploring common introductory Bayesian models: beta-binomial, gamma-Poisson, and normal-normal. Specifically, within these model settings, the bayesrules functions provide an active learning opportunity to interact with the three Bayesian model components, as well as the effects of different model settings on the model results. We present here the package’s visualization functions and how they can be utilized in a statistics classroom.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item