Program

Workshop Description:

This workshop assumes no prior experience with Bayesian statistical modeling, and is intended as both a theoretical and practical introduction. An understanding of Bayesian statistical modeling will be developed by relating it to participants' existing knowledge of traditional frequentist approaches. The philosophical underpinnings and departures from conventional frequentist interpretations of probability will be explained. This in turn will motivate the development of Bayesian statistical modeling.

To introduce Bayesian principles in familiar contexts we will begin with simple binomial and univariate normal models, then move to simple regression and multiple regression. Along the way, we will cover aspects of modeling including model construction, specifying prior distributions, graphical representations of models, practical aspects of Markov chain Monte Carlo (MCMC) estimation, evaluating hypotheses and data-model fit, model comparisons, and modeling in the presence of missing data. Although Bayesian statistical modeling has proven advantageous in many disciplines, the examples used in presentations draw primarily from social science and educational research. Examples will be accompanied by input and output from software. Throughout the course participants will be able to practice exercises using these software packages; participants are encouraged to bring their own laptops to perform these exercises. (Participants will be instructed on how to download free versions of the software prior to the course.)

Topics:

  • Machinery and Interpretations of Probability
  • Contrasting Frequentist and Bayesian Inference
  • Bernoulli/Binomial Models
  • Graphical Models
  • Accumulation of Evidence
  • Normal Distribution Models
  • Practical orientation to Markov chain Monte Carlo Estimation
  • Regression Models
  • Specifying Prior Distributions
  • Model Checking
  • Model Comparison
  • Missing Data Modeling

Prerequisites:

This workshop assumes no prior experience with Bayesian statistical modeling, and is intended as both a theoretical and practical introduction. Participants should have a foundational knowledge of conventional frequentist approaches to statistics (e.g., hypothesis testing, confidence intervals, least-squares and likelihood estimation) in contexts up through multiple regression. Although not required, a participant's experience in this workshop will be enhanced by additional prior coursework or experience with advanced statistical modeling techniques (e.g., general linear modeling, multivariate models for multiple outcomes) and/or by familiarity with the basics of probability theory (e.g., joint, marginal, and conditional distributions, independence).

Hardware & Software:

Models and exercises for these workshops will be done using the freely available R and JAGS software. Participants are welcome to bring their own laptops to perform these exercises, and will be instructed on how to download free versions of the software prior to the course.