DESCRIPTION

Package: clustra
Version: 0.2.1
Date: 2024-01-04
Title: Clustering Longitudinal Trajectories
Authors@R: c(person("George", "Ostrouchov", role = c("aut", "cre"), email = 
        "go@tennessee.edu"),
        person("David", "Gagnon", role = "aut"),
        person("Hanna", "Gerlovin", role = "aut"),
        person("Chen", "Wei-Chen", role = "ctb"),
        person("Schmidt", "Drew", role = "ctb"),
        person("Oak Ridge National Laboratory", role="cph"),
        person("U.S. Department of Veteran's Affairs", role="fnd",
        comment="Project: Million Veteran Program Data Core")
        )
Depends:
    R (>= 3.5.0)
Imports:
    data.table,
    graphics,
    grDevices,
    methods,
    mgcv,
    MixSim,
    parallel,
    stats
Suggests:
    haven,
    knitr,
    rmarkdown
Description: Clusters longitudinal trajectories over time (can be unequally 
    spaced, unequal length time series and/or partially overlapping series) on
    a common time axis. Performs k-means clustering on a single continuous 
    variable measured over time, where each mean is defined by a thin plate 
    spline fit to all points in a cluster. Distance is MSE across trajectory 
    points to cluster spline. Provides graphs of derived cluster splines, 
    silhouette plots, and Adjusted Rand Index evaluations of the number
    of clusters. Scales well to large data with multicore parallelism available
    to speed computation.
LazyLoad: yes
License: BSD 2-clause License + file LICENSE
Encoding: UTF-8
Maintainer: George Ostrouchov <go@tennessee.edu>
Roxygen: list(markdown = TRUE)
RoxygenNote: 7.2.3
URL: https://github.com/MVP-CHAMPION/clustra
BugReports: https://github.com/MVP-CHAMPION/clustra/issues
VignetteBuilder: knitr
LazyData: true
LazyDataCompression: xz