- Stata- Downloadable content available from our Stata Software page
- WinBUGS code and data
- Matlab code
- R Packages
- SAS Macros
- C code
WinBUGS 1.4 code and data
Supplementary WinBUGS 1.4 software and data to reproduce the base case analysis presented in the paper "Structural and parameter uncertainty in Bayesian cost-effectiveness models" (Jackson, C. H., Sharples, L. D., Thompson, S. G.)
Supplementary Material for paper;
"Making predictions from complex longitudinal data, with application to planning monitoring intervals in a national screening programme"
by Sweeting MJ and Thompson SG.
- MATLAB programs to fit the semiparametric non-stationary PACF model described in Su and Daniels (2013), 'Bayesian modelling of the covariance structure for irregular longitudinal data using the partial autocorrelation function'. Specifically:
- ACTG.m: the function to conduct MCMC for fitting the semiparametric non-stationary PACF model for the pediatric ACTG data
- nonpPACF.m: the main function to conduct MCMC
- MargCorr.m: function to convert a partial autocorrelation matrix to a marginal correlation matrix, using MATLAB build-in algorithiam for matrix inversion (created by Yanpin Wang, see Wang & Daniels (2013))
- MargCorr2.m: function to convert a partial autocorrelation matrix to a marginal correlation matrix, using the matrix inversion trick described in Wang & Daniels (2013) and Su and Daniels (2013)
- InvCorr.m: function to invert a correlation matrix with a corresponding band 'a' partial autocorrelation matrix, using the matrix inversion trick described in Wang & Daniels (2013) and Su and Daniels (2013)
- summaryMCMC: program to check convergence of the MCMC chains, create summary statistics and graphs, calculate DIC (integrated over random effects)
- "GUESS-ing polygenic associations with multiple phenotypes using a GPU-based Evolutionary Stochastic Search algorithm"
Leonardo Bottolo1, Marc Chadeau-Hyam, David I. Hastie, Tanja Zeller, Benoit Liquet, Paul Newcombe, ..., and Sylvia Richardson, In press Plos Genetic (2013)
To install R2GUESS package, simply type:
R CMD INSTALL R2GUESS_0.9.tar.gz
in the directory where the archive has been saved. Note that this package requires dependancies on several R package that first you have to install: snowfall, mixOmics, MCMCpack, mvtnorm, fields. To install these packages, simply type in a R console for example: install.packages("snowfall").
R2GUESS package is a wrapper of the GUESS (Graphical processing Unit Evolutionary Stochastic Search ) program. GUESS is a computationally optimised C++ implementation of a fully Bayesian variable selection approach that can analyse, in a genome-wide context, single and multiple responses in an integrated way. The program uses packages from the GNU Scientific Library (GSL) and offers the possibility to re-route computationally intensive linear algebra operations towards the Graphical Processing Unit (GPU) through the use of proprietary CULA-dense library.
Extensive documentation detailing the implementation of GUESS as well as all its features and options is available in http://www.bgx.org.uk/software/guess.html.
Note that R2GUESS is a beta version and a stable one will be soon released on CRAN.
- "Bayesian continuous reassessment method (CRM) designs for Phase I dose-finding trials"
This package implements a wide variety of one and two-parameter Bayesian CRM designs. The program can run interactively, allowing the user to enter outcomes after each cohort has been recruited, or via simulation to assess operating characteristics.
- "Individual patient data meta-analysis of time-to-event outcomes: one-stage versus two-stage approaches to estimating the hazard ratio under a random effects model."
J. Bowden, J. Tierney, M. Simmonds, A.J. Copas and J.P.T. Higgins. Research Synthesis Methods 2011.
- "Likelihood Estimation for a Longitudinal Negative Binomial Regression Model with Missing Outcomes."
Bond, SJ and Farewell VT
- "A random effect variance shift model for detecting outliers in meta-analysis."
Gumedze, F and Jackson, D. September 2010
- "Modelling multiple sources of dissemination bias in meta-analysis.
Jack Bowden, Dan Jackson, Simon G. Thompson. Statistics in Medicine, 2010. 29(7):945-955
- "ABSORB: A computer program for Assessing Bias using Sensitivity-analysis for Outcome Reporting Biases."
Dan Jackson. July 1, 2010
- "Multi-state Markov and hidden Markov models in continuous time."
See C. H. Jackson. Multi-State Models for Panel Data: The msm Package for R. Journal of Statistical Software, 2010. In press
- "Density strips and other methods for compactly illustrating distributions."
C. H. Jackson. Displaying uncertainty with shading. The American Statistician, 2008. 62(4):340-347.
- "Ecological regression using aggregate and individual data."
C. H. Jackson, N. G. Best and S. Richardson. Hierarchical related regression for combining aggregate and individual data in studies of socio-economic disease risk factors.Journal of the Royal Statistical Society, Series A: Statistics In Society, 2008. 171(1):159-178.
- "SAS code for the estimation and between-group comparison of cumulative incidence
functions in competing risks survival analysis"
Simon Bond. Pharmaceutical Programming. 2011. 4(1-2):1-13
- cuminc.sas: A macro to calculate the cumulative incidence curve for competing risks data
- gray.sas: A macro to calculate Gray's test comparing the cumulative incidence curves between groups
Gray RJ. A class of k-sample tests for comparing the cumulative incidence of competing risks.
Ann Stat. 1988; 16:1141-1154
- relabel.sas: A macro used by gray.sas to map a variable with an arbitrary set of values to a set of consecutive integers starting at 1
- "Identifying combined design and analysis procedures in two-stage trials with a binary end point."
Jack Bowden & James Wason. Statistics in Medicine. 2012. http://dx.doi.org/10.1002/sim.5468
Richard Nixon's Code Archive
Richard left the Biostatistics Unit early in 2007. Below is an archive of his downloadable code that is referenced in his publications.
- Download WinBUGS to use the code.
- Nixon RM and Thompson SG, Incorporating covariate adjustment, subgroup analysis and between-centre differences into cost-effectiveness evaluations, Health Economics, 2005
- Normal-normal [model] This file is the most commented - READ THIS FIRST [inital values] [prior distribution limits]
- Gamma-gamma [model] [inital values] [prior distribution limits]
- Gamma-gamma with covariate adjustment [model] [inital values] [prior distribution limits]
- Gamma-gamma with difference between subgroups [model] [inital values] [prior distribution limits]
- Gamma-gamma with fixed-effect difference between centres [model] [inital values] [prior distribution limits]
- Gamma-gamma with random-effects difference between centres [model] [inital values] [prior distribution limits]
- Thompson SG, Nixon RM and Grieve R, Addressing the issues that arise in analysing multicentre cost data, with application to a multinational study, Submitted to Journal of Health Economics, 2005
- Nixon RM, Wonderling D and Grieve R. How to estimate cost-effectiveness acceptability curves, confidence ellipses and incremental net benefits alongside randomised controlled trials. Submitted to Health Economics Letters, 2005
- Nixon RM Bansback N and Brennan A. Using mixed treatment comparisons and meta-regression to perform indirect comparisons to estimate the efficacy of biologic treatments in rheumatoid arthritis Submitted to Statistics in Medicine, 2006
- Nixon RM and Bansback N and Stevens JW and Brennan A and Madan J. How useful is short-term evidence in designing long-term clinical trials in rheumatoid arthritis?: Predicting ACR20 and ACR50 at six months from ACR20 and ACR50 at one and three months in clinical trials. Submitted to Arthritis Research and Therapy, 2006
- Nixon RM, Wonderling D and Grieve R. Non-parametric methods for cost-effectiveness analysis: the central limit theorem and the bootstrap compared. Submitted to Health Economics, 2007.
- RM Nixon, J Madan, SW Duffy, P Tappenden, J Chillcot, FH Cafferty. Assessing screening sensitivity and progression rates of colorectal cancer using multi-state modeling. Submitted to Biometrics, 2007.