FitsMEDseq models introduced by Murphy et al. (2021) <doi:10.1111/rssa.12712>,i.e. fits mixtures of exponential-distance models for clusteringlongitudinal/categorical life-course sequence data via the EM/CEMalgorithm. A family of parsimonious precision parameter constraints areaccommodated. So too are sampling weights. Gating covariates can besupplied via formula interfaces. Visualisation of the results of suchmodels is also facilitated.
The most important function in theMEDseq packageis:MEDseq_fit, for fitting the models via EM/CEM. Thisfunction requires the data to be in"stslist" format; thefunctionseqdef is conveniently reexported from theTraMineR package for this purpose.
MEDseq_control allows supplying additional argumentswhich govern, among other things, controls on the initialisation of theallocations for the EM/CEM algorithm and the various model selectionoptions.MEDseq_compare is provided for conducting modelselection between different results from using different covariatecombinations &/or initialisation strategies, etc.MEDseq_stderr is provided for computing the standard errorsof the coefficients for the covariates in the gating network.
A dedicated plotting function exists for visualising various aspectsof the results, using new methods as well as some existing methodsadapted from theTraMineR package. Finally, the packagealso contains two data sets:biofam andmvad.
You can install the latest stable official release of theMEDseq package from CRAN:
install.packages("MEDseq")or the development version from GitHub:
# If required install devtools: # install.packages('devtools') devtools::install_github('Keefe-Murphy/MEDseq')In either case, you can then explore the package with:
library(MEDseq) help(MEDseq_fit) # Help on the main modelling functionFor a more thorough intro, the vignette document is available asfollows:
vignette("MEDseq", package="MEDseq")However, if the package is installed from GitHub, the vignette is notautomatically created. It can be accessed when installing from GitHubwith the code:
devtools::install_github('Keefe-Murphy/MEDseq', build_vignettes = TRUE)Alternatively, the vignette is available on the package’s CRANpage.
Murphy, K., Murphy, T. B., Piccarreta, R., and Gormley, I. C. (2021).Clustering longitudinal life-course sequences using mixtures ofexponential-distance models.Journal of the Royal StatisticalSociety: Series A (Statistics in Society), 184(4): 1414–1451.<doi:10.1111/rssa.12712>.