Mixture
class MixtureBootstrapResults(val componentCountFrequency: IntegerFrequency, val familyByPosition: Map<Int, Map<String, Int>>, val modalComponentCount: Int?, val numSamples: Int, val numFitted: Int, val distinctValues: Statistic, val isParametric: Boolean, val originalDistinctValues: Int = 0)(source)
What repeated refitting on resampled data said about a recommendation.
Parameters
component
how often each number of components was recommended
family
for each component position, how often each family was chosen, counted only over resamples that recommended the modal number of components
modal
the most frequently recommended number of components
num
the resamples attempted
num
the resamples that produced a recommendation
distinct
the number of distinct values in each resample, which matters here because cuts may not split tied observations
is
whether resamples were drawn from the fitted mixture rather than from the data
original
how many distinct values the original data holds, so that the resamples' reduction can be seen rather than inferred