plot_pvalues: Function to plot p-values for different tests of hypothesis

In mosclust: Model Order Selection for Clustering

Function to plot p-values for different tests of hypothesis

Description

The p-values corresponding to different k-clusterings according to different hypothesis testing are plotted. A horizontal line corresponding to a given alpha value (significance) is also plotted. In the x axis is represented the number of clusters sorted according to the value of the stability index, and in the y axis the corresponding p-value. In this way the results of different tests of hypothesis can be compared.

Usage

plot_pvalues(l,alpha=1e-02,legendy=0, leg_label=NULL, colors=TRUE)

Arguments

l	a list of lists. Each component list represents a different test of hypothesis, and it has in turn 4 components: ordered.clusterings : a vector with the number of clusters ordered from the most significant to the least significant; p.value : a vector with the corresponding p-values computed according to chi-square test between multiple proportions in descending order (their values correspond to the clusterings of the vector ordered.clusterings); means : vector with the mean similarity (stability index) for each clustering; variance : vector with the variance of the similarity for each clustering.
alpha	alpha value for which the straight line is plotted
legendy	ordinate of the legend. If 0 (def.) no legend is plotted.
leg_label	labels of the legend. If NULL (def.) the text "test 1, test 2, ... test n" for the n tests is printed. Otherwise it is a vector of characters specifying the text to be printed
colors	if TRUE (def.) lines are printed with colors, otherwise using only different line pattern

Value

No return value, the function is called for its side-effect of drawing a plot on the current graphics device.

Author(s)

Giorgio Valentini [e-mail]

See Also

plot_cumulative.multiple

Examples

library("clusterv")
# Data set generation
M <- generate.sample6 (n=20, m=10, dim=1000, d=3, s=0.2);
# generation of multiple similarity measures by resampling
Sr.kmeans.sample6 <- do.similarity.resampling(M, c=10, nsub=20, f=0.8, s=sFM, 
                                      alg.clust.sim=Kmeans.sim.resampling); 
# hypothesis testing using the chi-square based test
d.chi <- Chi.square.compute.pvalues(Sr.kmeans.sample6)
# hypothesis testing using the Bernstein based test
d.Bern <- Bernstein.compute.pvalues(Sr.kmeans.sample6)
# hypothesis testing using the Bernstein based test (with independence assumption)
d.Bern.ind <- Bernstein.ind.compute.pvalues(Sr.kmeans.sample6)
l <- list(d.chi, d.Bern, d.Bern.ind);
# plot of the corresponding computed p-values
plot_pvalues(l, alpha = 1e-05, legendy = 1e-12)

mosclust documentation built on June 8, 2025, 11:23 a.m.