Introduction to STA 360-602

Rebecca C. Steorts
17 January 2017

Instructor and TA's

  • Professor Rebecca C. Steorts, Assistant Professor of Statistical Science and Computer Science, Office Hours: Tu: 1–2 PM, Old Chem 216, beka@stat.duke.edu

  • Lei Qian, Office Hours: Thursday: 9:00–11:00 AM, Old Chem 211A, lsq3@duke.edu

  • Derek Owens-Oas, Office Hours: Wednesday, 6:00–8:00 PM, Old Chem 211A derek.owens.oas@duke.edu

  • Xingxu Yan, Office Hours: Monday, 4:30–6:30 PM, Old Chem 211A xingyu.yan@duke.edu

Questions + Resources

  • All questions (especially of length) should occur in office hours
  • Quick questions or clarifications, should occur via the Google group https://groups.google.com/forum/#!forum/bayes17
  • Readings, office hours are on the Google group
  • Assignments will be posted to Sakai along with due date
  • Announcements will be sent via Google group or email

Course Structure

  • Introduction to R
  • Introduction to Bayesian Statistics
  • Decision Theory
  • Hierarchical Models
  • Monte Carlo
  • Markov Chain Monte Carlo (MCMC)
  • Gibbs Sampling
  • Multivariate Bayesian Models
  • Linear Regression
  • Special Topics

Lectures and Lab

  • Lecture slides will be posted before lecture.
  • Please attend lab. Part of your homeworks will be based upon lab.
  • Office hours: see the syllabus.
  • Google group.
  • Required readings (see the Google group and updates).
  • When emailing the professor or a TA, please CC Prof Steorts and TAs for the fastest response.

Labs

  • Will correspond to the material from lecture
  • You will complete a few tasks with the TAs (ungraded)
  • You will finish the tasks as part of your homework assignment (graded)

Lecture

  • Typos: Please note these and give them to me at the end of lecture.
  • Questions: Will take these at the end of the module for 10 minutes. All other questions should be addressed in office hours.
  • Practice Problem: There will be one practice problem every lecture. These will serve as practice problems for the exams.

Homeworks

  • You may work together on homeworks, but plagirism of any form (including copying of code) will not be tolerated.
  • All homeworks must be submitted as one .pdf file. Please include all source files that you work with as well. Recommend: zip all files and add these to Sakai.
  • Please write legibly for all proofs or written problems. Include all work for full credit.
  • For applied problems, be sure to include your markdown file, make sure it is properly documented, and that it runs, and include the final .pdf file with results and commentary.
  • Late homeworks will not be accepted. Your lowest homework will be dropped.
  • 10 pts for clarity, commenting of code, writing/explanations.

Exams

  • In class
  • Closed-book, closed-notes
  • Based upon lectures, labs, homeworks, and practice problems
  • No make up exams
  • You must attend the final exam to pass the course

Grades

  • Homework - 25 %, expect about 8 total assignments for the semester
  • Exam I - 30 %, Feb 9
  • Exam II - 30 % March 9
  • Final Exam (comprehensive) - 15 %
  • You may ask Professor Steorts in person for your grade and class ranking at any time after Exam I.