140.763.01
Bayesian Methods II
Location
East Baltimore
Term
4th Term
Department
Biostatistics
Credit(s)
3
Academic Year
2025 - 2026
Instruction Method
In-person
M, W, 1:30 - 2:50pm
Auditors Allowed
Yes, with instructor consent
Available to Undergraduate
No
Grading Restriction
Letter Grade or Pass/Fail
Course Instructor(s)
Contact Name
Frequency Schedule
Every Other Year
Resources
Prerequisite
• 140.776 or equivalent
• 140.777 or familiarity with rmarkdown, knitr, and Git(Hub)
• 140.651–654 or 140.751–754, with completion of 140.762 (Bayesian Methods 1) is highly encouraged for those who have completed 651–654 rather than 751–754
• 140.646–649 or 140.721–724
• (Recommended) 140.778 and 140.779, or familiarity with Monte Carlo and optimization methods
Enrollment Restriction
This course is not restricted.
Serves as a sequel or more advanced alternative to Bayesian Methods I (140.762), proceeding more quickly than a typical first course on Bayesian statistics by building on knowledge of classical statistics but otherwise assuming no prior exposure to Bayesian paradigms. Introduces Bayesian concepts and machinery of general interest regardless of whether one subscribes to Bayesian philosophy, exploring similarities to and differences from frequentist paradigms whenever appropriate. Discusses Bayesian analyses and theories of: classical parametric models; shrinkage estimation and regularized/penalized methods; hierarchical/mixed-effect regression models; hyperparameter and model selections; and non-parametric methods. Familiarizes students with computational techniques and software packages to fit Bayesian models.
Learning Objectives
Upon successfully completing this course, students will be able to:
- Articulate similarities and differences between Bayesian and frequentist approaches
- Identify when (and when not) to apply Bayesian methods in real data analyses
- Draw on Bayesian thinking to derive statistical procedures with well-calibrated frequentist properties, when devising such procedures is difficult otherwise
- Deploy common Bayesian modeling techniques such as hierarchical and latent variables models in a variety of applications
- Learn and identify more advanced Bayesian methods as needed for particular applications
Upon successfully completing this course, students will be able to:
Methods of Assessment
This course is evaluated as follows:
- 80% Homework
- 20% Final Presentation