Lawrence Joseph
Bayesian statistics

Please Note: Prof. Joseph has retired.
These pages are left up in case they prove
useful, but the pages and software will
no longer be updated. All material and
software is "as is" with no guarantees
of functionality or correctness.
Basic info
Course Outline
1 Mon Jan 9 Introduction/Motivation/Evaluation/Scope
2 Wed Jan 11 Multivariate Distributions, Conditionality
3 Mon Jan 16 Basic Elements of Bayesian Analysis
4 Wed Jan 18 Bayesian Philosophy I
  James Berger - Example 8
James Berger - Example 13
James Berger - Example 17
Berger and Berry - Illusion of Objectivity
Dunson - Practical Advantages of Bayes in Epidemiology
5 Mon Jan 23 Bayesian Philosophy II
6 Wed Jan 25 Simple Models I - Univariate Models
  R Lecture 1
R Lecture 2
R Lecture 3
R Lecture 4
R program for Normal Means - Version 1
R program for Normal Means - Version 2
R program for exact binomial confidence intervals
7 Mon Jan 30 Simple Models II - Predictive Distributions
8 Wed Feb 1 Computation and Numerical Methods I - Introduction
9 Mon Feb 6 Computation and Numerical Methods II - Monte Carlo Integration
10 Wed Feb 8 Computation and Numerical Methods III - SIR Algorithm
  SIR program #1
SIR program #2
SIR program #3
11 Mon Feb 13 Computation and Numerical Methods IV - Gibbs sampler and WinBUGS
  Simple Gibbs program in R
Paper on adaptive rejection sampling
12 Wed Feb 15 Computation and Numerical Methods V - More on WinBUGS
  WinBUGS Quick Reference for Model Preparation
WinBUGS Quick Reference for Analysis
WinBUGS program for Binomial Proportion
WinBUGS program for Binomial Proportion Difference
WinBUGS program for Normal Mean, Known Variance
WinBUGS program for Normal Mean, Unknown Variance
WinBUGS program for Linear Regression
WinBUGS program for Logistic Regression
WinBUGS program for Hierarchical Binomial Proportion
All WinBUGS programs in one pdf file
13 Mon Feb 20 No Class -- Spring Break
14 Wed Feb 22 No Class -- Spring Break
15 Mon Feb 27 Bayesian Linear and Logistic Regression
16 Wed Feb 29 Hierarchical Linear and Logistic Regression
  WinBUGS rats example
WinBUGS seeds example
WinBUGS pumps example
17 Mon Mar 5 Bayesian Analysis of Clinical Trials
  Brophy and Joseph - Placing trials in context using Bayesian analysis
Hughes 1993
Speigelhalter et al 1994
Fayers et al 1997
Fisher 1999
18 Wed Mar 7 Hierarchical Models I - Simple Hierarchical Models
19 Mon Mar 12 Hierarchical Models II - Meta Analysis with Random Effects
  Smith 1994
Carlin 1992
WinBUGS blocker example
20 Wed Mar 14 Hierarchical Models III - More Complex Hierarchical Models
  WinBUGS orange tree example
WinBUGS inhalers example
WinBUGS kidney example
Brophy 1993
Accompanying program to Brophy 1993
21 Mon Mar 19 Adjusting for Measurement Error
  Richardson and Gilks
WinBUGS Air Example
WinBUGS cervix example
22 Wed Mar 21 Prior Distributions - Prior Selection and Elicitation
  Chaloner 1994
23 Mon Mar 26 Model Selection in Regression - Bayes Factors
  Bayes Factor graphical explanation
Raftery 1995
Kass 1995 (abstract only)
WinBUGS pines example
24 Wed Mar 28 Missing Data
  Real example of missing data with non-ignorable missingness
25 Mon Apr 2 Bayesian bias adjustments
  Greenland 2005
26 Wed Apr 4 Bayesian Sample Size Criteria
  Moore 1999
Joseph 1997
27 Mon Apr 9 No Class -- Easter Monday
28 Wed Apr 11 Analysis of Diagnostic Test Data
  Plain text version of slides
Joseph 1995
Diagnostic test program for one test
Diagnostic test program for two tests
Diagnostic test program for three tests
Example of use of tt2 program
29 Mon Apr 16 Discussion and Conclusions - The Future of Bayesian Analysis
  Berger article and discussion notes
Top 10 reasons to be Bayesian