ST740 -- Bayesian Inference -- Fall 2011

---

Course times/location: MW 10:15-11:30, 1108 SAS Hall
Instructor: Brian Reich (brian_reich@ncsu.edu)
Office hours: 4262 SAS Hall, MW 1-2pm
Syllabus: http://www4.stat.ncsu.edu/~reich/st740/Syllabus

Homework assignments

---

• HW 1 (Due: 8/31)
  
• HW 2 (Due: 9/14)
  
• HW 3 (Due: 10/5) Adam's solutions
  
• HW 4 (Due: 10/24)
  
• Exam 2: description and Data
• HW 5 (Due: 11/30)
• Final Project

Course outline

---

Week -- Description
8/17   -- Introduction
8/22   -- Bayes Basics
8/29   -- Priors
9/7     -- Empirical Bayes; Labor Day (M)
9/12   -- Bayesian performance
9/19   -- MCMC; Midterm Exam (W, 9/21)
9/26   -- MCMC
10/3   -- MCMC
10/10 -- MCMC diagnostics
10/17 -- Linear/Generalized linear models
10/24 -- Hierarchical models
10/31 -- Semiparametric Bayes; Midterm Exam (W, 11/2)
11/7   -- Variable selection
11/14 -- Model selection
11/21 -- Model diagnostics; Thanksgiving (W)
11/28 -- Special topics
12/7   -- Final exam, Wednesday, 8-11am, 1108 SAS Hall

Computing/data/references

---

• R code to the plot the prior/posterior for the beta/binomial model
• Scottish lip cancer data and analysis code
• Gelman A (2006). Prior distributions for variance parameters in hierarchical models. Bayesian Analysis , 1, 515-533.
• Liu A, Hodges JS (2003). Posterior bimodality in the balanced one-way random-effects model. JRSS-B , 65, 247-255.
• WinBUGS code for Poisson regression.
• Hiance A, Chevret S, Levy V (2009). A practical approach for eliciting expert prior beliefs about cancer survival in phase III randomized trial. Journal of Clinical Epidemiology, 62, 431-437.
• R code to simulate an emprical Bayes analysis of a normal mean.
• Gelman A (2008). Objections to Bayesian statistics. Bayesian Analysis , 3, 445-450.
• R code to the plot the bias, variance, and MSE for the beta/binomial model
• R code to compare interval estimates for the binomial proportion as in Agresti and Coull (TAS, 1998).
• R code to perform univariate rejection sampling
• An introduction to INLA
• MCMC code for the Bayesian t-test
• MCMC code for the Bayesian linear model
• MH code for the Bayesian logistic regression model
• Description of SAS Proc MCMC.
• Sample OpenBUGS code.
• R code to call OpenBUGS to analyze this model.
• Code to explore MCMC diagnostics.
• Charlie Geyer's thoughts on MCMC diagnostics including multiple chains and burn-in.
• A Bayesian Nobel recipient and his views on Bayesian Methods in Applied Econometrics
• BUGS code for a linear model with missing data.
• Probit regression code.
• BUGS code for longitudinal data.
• R code for spatial binary data, along with adjacency matrix of NC counties and 2008 Presidential election results.
• Draw samples from a mixture of normals.
• R code for linear regression with a mixture of normals model for the residuals.
• Semiparametric regression for the motorcycle data.
• WinBUGS code for a stochastic search variable selection.
• R code for a stochastic search variable selection.
• R code for a DIC analysis of the motorcycle data.
• R code to compute DIC for the one-way random effects model.
• Predictive diagnostics using OpenBUGS.
• What are the open problems in Bayesian statistics? By Michael Jordan.
• Berry DA (2006). Bayesian clinical trials. Nature Reviews, 5, 27-36.
• Bayesian sample size calculations.