Stat 151: Intro to Bayesian Statistics
Content will be available on the course Learning Suite webpage.
Course Description and Learning Outcomes
This course serves as an introduction to statistics from the Bayesian paradigm (instead of the more traditional introduction to statistics from a frequentist paradigm). Topics include the scientific method, conditional probability, Bayes' Theorem, Gibbs sampling, and conjugate distributions such as the Beta-binomial, Poisson-gamma, and normal-normal.
- Explain how conditional probability and Bayes' Theorem relate to the analysis of data via the Bayesian paradigm
- Identify the conjugate priors of the normal (mean and variance), binomial, and Poisson distributions and derive the respective posterior distributions
- Explain why Gibbs and Metropolis samplers work and when they are appropriate to use
- Code in R a Gibbs sampler and/or Metropolis sampler for a simple non-conjugate posterior distributions
- Interpret and explain the results of Bayesian analysis