Dr Adrian Barnett
E-mail: a.barnett@uq.edu.au. My CV (updated October 2007).
I am leaving UQ to join IHBI at QUT in November 2007.
I graduated from University College London with a BSc in statistics
in 1994. After that I worked for SmithKline Beecham and the Medical
Research Council as a statistician before coming to Australia
to do a PhD. I completed my PhD in time series in 2002, for more
details on the project see http://adt.library.uq.edu.au/public/adt-QU20021203.122401/.
My main interest is in the application of statistics to health. I am particularly interested in problems involving longitudinal or spatial data. I have worked on studies of air pollution (currently working on the ACHAPS study) and cardiovascular disease (WHO MONICA project). I am interested in Bayesian methods, non-linear analysis and bootstrap methods.
Students:
Ignacio Correa (PhD, Associate Supervisor), graduated July 2004
Jo-An Atkinson (Masters, Associate Supervisor), graduated December 2005
Darren Gray (PhD, Associate Supervisor)
Letty Burridge (PhD, Associate Supervisor)
Kathryn Plonka (PhD, Associate Supervisor)
Presentations and statistical code:
The rise of Bayesian statistics in Epidemiology: web page.
Zipped
SAS code. to run the Bland-Altman method for multiple observations per individual. (Paper: Bland JM, Altman D. (2007) Agreement between methods of measurement with multiple observations per individual. Journal of Biopharmaceutical Statistics 17, 571–582). See also Martin Bland's homepage.
Slides from the 2006 Spring Bayes meeting: Zipped
slides.
SAS macro to perform time-stratified case-crossover analysis
with air pollution example from the NMMAPS
study: Zipped
files.
Zipped code and datasets from a one-day course on GEEs: Web
page.
Zipped MATLAB m-files of a test of non-linearity published
in the IEEE Transactions on Signal Processing (2005, vol 53, issue
1): Zipped
m-files.
Zipped SAS files of a test of seasonality published in Statistics
in Medicine (2004, vol 23, issue 22): Zipped
SAS files (updated July 2006). The programs require the SAS modules of STAT, GPLOT,
ETS and IML. For a practical example of the method using the WHO
MONICA data click here.
Zipped CSV and SAS data sets from Diggle et al. 'Analysis of
Longitudinal Data': Zipped
data sets (not complete).
Zipped PowerPoint slides (with hidden SAS code) from my presentation
on Chapter 3 of Diggle et al.'s book 'Analysis of Longitudinal
Data': Zipped
slides.
Zipped PowerPoint slides from my presentation to the Brisbane
Bayes Interest Group on Chapter 7 of Peter Congdon's book 'Bayesian
Statistical Modelling': Zipped
slides.
SAS code to draw classification trees: SAS
code.
R code to run a classification tree with an example data set:
Web
page.
Regression to the mean SAS code with an example data set: Web
page.
My poster at the International Society for Environmental Epidemiology
conference, 2004: Powerpoint
slide. |
|