All course materials for the "Statistical Modelling in Stata"
course can be found here
Guide to Propensity Analysis
A guide to performing propensity analyses in Stata can be found here. It covers how to
calculate propensity scores, check that they work, and use them to
estimate the effect of an exposure. The do-file used to perform
all of the analyses is available as an appendix in the guide, or
you can download it here if that
is more convenient.
Guide to Multiple Imputation
A guide to multiple imputation, and the dataset used for the
examples in the guide can he found here. This is of particular interest if you
need to impute continuous variables with a non-normal
Tutorial in competing risks
This practical aims to
illustrate some of the problems caused by competing risks in
Survival Analysis, and present some of the solutions available in
Stata. It is based on "Tutorial in biostatistics: Competing risks
and multi-state models" by Putter H, Fiocco M, and Geskus RB,
published in Statistics in Medicine 2007 volume 26, pages
2389--2430. The original tutorial used R for the analysis, this
uses stata instead. A do-file to perform the entire analysis can
be found here.
Guide to using the Linux Workstation
Some notes about connecting to the
Linux workstation from a PC. Only of interest to people within the
ARC epidemiology unit, who could be given access to it.
I have written some software for Stata. To view it, start stata
net from http://personalpages.manchester.ac.uk/staff/mark.lunt
in the command window.
Alternatively, all the files are available
here. However, the help files are
written in stata's own SMCL language, so are not easy to read
other than in stata. The ado files should be fairly
comprehensible to someone with a good knowledge of the C
programming language and some familiarity with statistical
concepts, although some knowledge of the peculiarities of
stata's own programming language would be very helpful.