CSSS-Stat 564 Bayesian Statistical Methods (Spring 2012)

Lectures TTh, 10:30-11:45,  RAI 116
Labs T, 3:30-4:20, Communications B027
R and JAGS code seen in lectures 

Instructor
  • Youyi Fong
  • Padelford C-14E
  • Office Hours: T/Th 12:30-1:30 or by appointment
  • yfong@uw.edu
 Teaching Assistant
  • Wen Wei Loh
  • Padelford C-312
  • Office Hours: Tues 4:30-5:30 and Wed 3:00-4:00
  • wloh@uw.edu


Please include "564" (without quotes) in any emails to allow for appropriate filtering.
Texts/Resources
Assignments
  • Week 10:
    • Read Chapter 11 and 12
  • Week 9:
    • Read Chapter 11
    • HW8, due Thu, May 31
      • 11.2 (a) to (d). In (b), use 5 degrees of freedom for Wishart. In (c), use JAGS to do the sampling. Remember to center every covariate. In (d), just plot the diagonal elements of Sigma.
      • Extra credit 11.2 (e)
  • Week 8:
    • Read Chapter 9, 10
    • HW 7, due Thu, May 24
      • 10.2 (d), (e). Use JAGS to perform posterior sampling. Try two different priors as we did for the Poisson regression example.
        • default prior
        • 'unit information' prior
  • Week 7:
    • Read Chapter 8,
    • HW 6, due Thu, May 17
      • 8.1. 
      • 8.3, (a), (b), (c) and (e). You can use JAGS to do Gibbs sampling in (a) or write your own Gibbs code, extra credit for doing both.
  • Week 6 (April 29):
    • Read Chapter 6, 7
    • HW5, due Thu, May 10
      • 6.1 (a) and (d). Also draw a graphical model. In (d), use JAGS to do the sampling. 
      • 7.4. For part (d), only do d.iii, draw and graphical model and use JAGS to do the sampling.
  • Week 5:
    • Read Chapter 5 and 6
    • HW4, due Thu, May 3
      • 5.2
      • This week's homework is light because your one-page project summary is due soon on May 8. Please come talk to me if you have any questions.
      • Extra credit: 5.1
  • Week 4:
    • Read Chapter 4 and 5
    • HW3, due Thu, April 26
      • 3.3 part (a), 4.2. For 4.2 (b) use n0 values between 1 and 10. For 4.2 (c) you only have to repeat part 4.2 (a). 
      • 4.8 (a)-(b)
      • Extra credit: 4.8 (c)-(d)
  • Week 3: 
    • Read Chapter 3 and 4 
    • Do exercises 3.2, 3.7 (also draw a graphical model and write a JAGS model file ); extra credit: 3.9 (a)-(c). Due Thu, April 19. 
  • Week 2: 
  • Week 1: Read Chapter 1 and 2 of the text


Evaluation
  • Eight homework assignments
  • Final project, count as two homework assignments

Course Outline

  1. Introduction to Bayesian statistics
  2. Concepts of randomness and probability, review of probability calculus
  3. Inference for binomial and Poisson distributions
  4. Monte Carlo
  5. Inference for normal distribution
  6. Hierarchical models
  7. Multivariate normal distribution
  8. Linear regression models
  9. Generalized linear models
  10. Generalized linear mixed-effects models

Late policy 
Each turned in item receives an initial grade of x, then the actual grade is y=x exp(-d/8), where d is the number of days (including weekends) after the due date I receive the work. Everyone receives one grace period to be applied to one homework for the entire quarter.


Notes

4/12/12

  • Final project planning
    • One page summary of background, data and model due on May 8 
    • Meet to discuss projects in half an hour blocks during 12:30-3:30 on May 8 or 12:30-1:30 on May 10
    • 15-20 min class presentations and a report (2-3 page text + up to 3 tables and/or figures) due on May 31

4/10/12