plot.fclust

Plot method for the 'fclust' objects returned by <code>fclustering</code>.

Implementation of global envelopes for a set of general d-dimensional vectors T
in various applications. A 100(1-alpha)% global envelope is a band bounded by two
vectors such that the probability that T falls outside this envelope in any of the d
points is equal to alpha. Global means that the probability is controlled simultaneously
for all the d elements of the vectors. The global envelopes can be used for graphical
Monte Carlo and permutation tests where the test statistic is a multivariate vector or
function (e.g. goodness-of-fit testing for point patterns and random sets, functional
analysis of variance, functional general linear model, n-sample test of correspondence
of distribution functions), for central regions of functional or multivariate data (e.g.
outlier detection, functional boxplot) and for global confidence and prediction bands
(e.g. confidence band in polynomial regression, Bayesian posterior prediction). See
Myllymäki and Mrkvička (2024) <doi:10.18637/jss.v111.i03>,
Myllymäki et al. (2017) <doi:10.1111/rssb.12172>,
Mrkvička and Myllymäki (2023) <doi:10.1007/s11222-023-10275-7>,
Mrkvička et al. (2016) <doi:10.1016/j.spasta.2016.04.005>,
Mrkvička et al. (2017) <doi:10.1007/s11222-016-9683-9>,
Mrkvička et al. (2020) <doi:10.14736/kyb-2020-3-0432>,
Mrkvička et al. (2021) <doi:10.1007/s11009-019-09756-y>,
Myllymäki et al. (2021) <doi:10.1016/j.spasta.2020.100436>,
Mrkvička et al. (2022) <doi:10.1002/sim.9236>,
Dai et al. (2022) <doi:10.5772/intechopen.100124>,
Dvořák and Mrkvička (2022) <doi:10.1007/s00180-021-01134-y>,
Mrkvička et al. (2023) <doi:10.48550/arXiv.2309.04746>, and
Konstantinou et al. (2024) <doi: 10.1007/s00180-024-01569-z>.

Mari Myllym<c3><a4>ki

Global Envelopes

Mari Myllymäki

Tomáš Mrkvička

Mikko Kuronen

Jiří Dvořák

Pavel Grabarnik

Ute Hahn

Michael Rost

Henri Seijo

plot.fclust function

<dl><dt>x</dt>
<dd>An 'fclust' object.</dd>
<dt>plotstyle</dt>
<dd>The resulting central regions of clusters can be plotted by sorting the
appropriate <code>curve_set</code> only 'marginal' or by sorting the joined list of <code>curve_set</code>
objects 'joined'. If 'joined' is used the shown central regions corresponds to the joined
ordering used to cluster the functional data. If 'marginal' is used the shown central regions
do not correspond to the joined ordering used to cluster the functional data, but better express
the shape of cluster with respect to given <code>curve_set</code>.</dd>
<dt>coverage</dt>
<dd>The coverage of central regions to be used to show the clusters.</dd>
<dt>nstep</dt>
<dd>1 or 2 for how to contruct a combined (joined) global envelope
if there are more than one sets of curves. Default to 1, if the numbers of points
where the curves are observed (r) are the same in each set, and 2 otherwise.</dd>
<dt>ncol</dt>
<dd>The number of columns in the graphical output, when there is just one set of curves
that has been ordered. If not given, <code>c(1, k+1)</code> is used, which gives all plots in one row.
For more sets of curves, the rows are fixed to correspond to the sets (one row for each set).</dd>
<dt>...</dt>
<dd>Ignored.</dd></dl>

Arguments

Plot method for the class 'fclust' — plot.fclust

<dl>

<dt>x</dt>
<dd>An 'fclust' object.</dd>


<dt>plotstyle</dt>
<dd>The resulting central regions of clusters can be plotted by sorting the
appropriate <code>curve_set</code> only 'marginal' or by sorting the joined list of <code>curve_set</code>
objects 'joined'. If 'joined' is used the shown central regions corresponds to the joined
ordering used to cluster the functional data. If 'marginal' is used the shown central regions
do not correspond to the joined ordering used to cluster the functional data, but better express
the shape of cluster with respect to given <code>curve_set</code>.</dd>


<dt>coverage</dt>
<dd>The coverage of central regions to be used to show the clusters.</dd>


<dt>nstep</dt>
<dd>1 or 2 for how to contruct a combined (joined) global envelope
if there are more than one sets of curves. Default to 1, if the numbers of points
where the curves are observed (r) are the same in each set, and 2 otherwise.</dd>


<dt>ncol</dt>
<dd>The number of columns in the graphical output, when there is just one set of curves
that has been ordered. If not given, <code>c(1, k+1)</code> is used, which gives all plots in one row.
For more sets of curves, the rows are fixed to correspond to the sets (one row for each set).</dd>


<dt>...</dt>
<dd>Ignored.</dd>

</dl>

plot.fclust: Plot method for the class 'fclust'

Description

Usage

Arguments

Details

References