Statistical Software

It is frightful that someone who is no one ... can set any error into circulation with no thought of responsibility and with the aid of this dreadful disproportioned means of communication, Søren Kierkegaard's Journals and Papers, Edited and translated by H. V. Hong et. al., Vol. 2, p 481, 1967.

[R-forge] | [github]

Temporal and Spatio-Temporal Modelling and Monitoring

surveillance -- An R-package implementing prospective surveillance algorithms and containing functionality for the temporal and spatio-temporal modelling of epidemic phenomena. Seen in a broader statistical context the package implements retrospective modeling and prospective change-point detection in time series of counts, proportions and categorical data. Furthermore, the package implements function to perform statistical modelling of epidemic phenomena seen as multivariate counting processes or spatio-temporal point-processes.

Statistical Methods in Epidemic Modelling

RLadyBug -- An R package for the visualisation and inference in stochastic Susceptible-Exposed-Infectious-Recovered (SEIR) models and its spatio-temporal extensions. The package used to be available from CRAN.

LadyBug 2.0 -- MCMC estimation of partially observed epidemics implemented in Java using the Hydra library. This is an implementation of the theory presented in Inference in disease transmission experiments using stochastic epidemic models, M. Höhle, E. Jørgensen, and P.D. O'Neill, Journal of the Royal Statistical Society, Series C, volume 54, part 2 (2005), pages 349-366.

R-code in connection with study group and seminar on the book Stochastic Epidemic Models and Their Statistical Analysis, H. Andersson and T. Britton (2000), Lecture Notes in Statistics 151, Springer. The file README.html contains a more detailed description of the files useful for simulating SIR epidemics, calculating final sizes and estimating model parameters.

R-code to calculate R0 by the method of moments (i.e. according to Equation 7.4.3 in N. G. Becker's Analysis of Infectious Disease Data, Monographs on Statistics and Applied Probability. Chapman and Hall, 1989).
Data from the article Experimental quantification of vaccine induced reduction in virus transmission, M.C.M. De Jong and T.G. Kinman, Vaccine, 12:761–766, 1994 and the data used in the LadyBug article are given as example.

I contribute to the R-Epi project, which lists R resources for infectious disease epidemiology.

Population Size and Sample Size Calculations

binomSamSizeis an R package containing functions to compute confidence intervals and necessary sample sizes for the parameter p of the Bernoulli distribution under simple random sampling or under pooled sampling, e.g. when investigating prevalence in populations. Sample size calculations are based on expected length. The development version can be installed from github:library("devtools")
install_github("mhoehle/binomSamSize")

Maple and R-code accompanying the paper "Bayesian Estimation of the Size of a Population". A population of unknown size N is labeled using the serial numbers {1,2,...,N}. A random sample (X_1,...,X_n) of size n<N is observed without replacement. The task is to make Bayesian inference on the population size N.

Stochastic Petri Nets

Stochastic Petri Nets (SPNs) in R. The petRi (v1.0) package is a small R-package to read Petri Nets specified by CPN Tools and enhance them to SPNs. Supported features are: simulation (the token game), estimation of stationary distribution, analysis of mean number of tokens, distribution of firing transition, etc. Two examples from systems biology are given.

[petRi_1.0.tar.gz] [petRi_1.0.zip]

Computer Assisted Learning

Student programmer on the LDIP (digital photogrammetry) and LDIPInter (Image processing & digital photogrammetry) projects.
Note: The LDIP program won the 1996 International Society for Photogrammetry and Remote Sensing (ISPRS) - Computer Assisted Teaching CONtest (CATCON) Gold Award.