Skip Navigation

Course Directory

140.762.01
Bayesian Methods I

Location:
East Baltimore
Term:
3rd term
Department:
Biostatistics
Credits:
3 credits
Academic Year:
2022 - 2023
Instruction Method:
Synchronous Online with Some Asynchronous Online
Class Times:
  • Tu Th,  1:30 - 2:50pm
Auditors Allowed:
Yes, with instructor consent
Undergrads Allowed:
No
Grading Restriction:
Letter Grade or Pass/Fail
Course Instructors:
Contact:
Gary Rosner
Frequency Schedule:
Every Other Year
Next Offered:
2024 - 2025
Resources:
Prerequisite:

Biostatistics 140.651 and 140.652, or instructor consent

Description:

Illustrates current approaches to Bayesian modeling and computation in statistics. Describes simple familiar models, such as those based on normal and binomial distributions, to illustrate concepts such as conjugate and noninformative prior distributions. Discusses aspects of modern Bayesian computational methods, including Markov Chain Monte Carlo methods (Gibbs' sampler) and their implementation and monitoring. Bayesian Methods I is the first term of a two term sequence. The second term offering, Bayesian Methods II (140.763), develops models of increasing complexity, including linear regression, generalized linear mixed effects, and hierarchical models.

Learning Objectives:

Upon successfully completing this course, students will be able to:

  1. Explain the difference between the Bayesian approach to statistical inference and other approaches
  2. Develop Bayesian models for combining information across data sources
  3. Write and implement programs to run analyses
  4. Evaluate the influence of alternative prior models on posterior inference
  5. Plot and interpret posterior distributions for parameters of scientific interest
Methods of Assessment:

This course is evaluated as follows:

  • 75% 6 homework assignments, each worth 12.5%
  • 25% Final Exam

Instructor Consent:

No consent required