Bayesian Statistics: Techniques and Models

This module provides an introduction to statistical modeling and Monte Carlo estimation. It covers the modeling process, components of Bayesian models, non-conjugate models, Monte Carlo integration, Markov chains, and the statistical modeling process. Learners will gain insights into efficient communication of data analysis results and statistical modeling techniques.

The Markov chain Monte Carlo (MCMC) module delves into the algorithm, random walk examples, model writing, running, and post-processing, trace plots, autocorrelation, and alternative MCMC software. Participants will gain practical knowledge of MCMC methods and their application in statistical modeling and analysis.

This module introduces common statistical models, including linear regression, ANOVA, and logistic regression. It covers model setup in R, JAGS model development, model checking, DIC, and prediction. Learners will understand the application of these models in analyzing scientific questions involving various types of data.

The count data and hierarchical modeling module explores Poisson regression, predictive distributions, correlated data, prior predictive simulation, hierarchical modeling, and mixture models. Participants will learn to select prior distributions and apply hierarchical modeling to analyze complex data.

The capstone project module concludes the course with a comprehensive data analysis project that allows learners to apply the knowledge and skills acquired throughout the course. Participants will engage in a hands-on project to further reinforce their understanding of Bayesian statistical models and computational techniques.