DataVisualizations (version 1.1.4)

MDplot: Mirrored Density plot (MD-plot): Visualization for a Boxplot-like Shape of the PDF

Description

This function creates a MD-plot for each variable of the data matrix. The MD-plot is an improvement of violin or so-called bean plots which posses advantages in comparison to the conventional well-known box plot.

Usage

MDplot(Data, Names,fill='darkblue',scale='width',size=0.01)

Arguments

Data

Matrix containing the data. Each column is one variable.

Names

Optional: Names of the variables. If missing the columnnames of data are used.

fill

Optional: color with which violin is to be filled with, see ggplot2 documentation for details

scale

Optional: if "area" (default), all violins have the same area (before trimming the tails). If "count", areas are scaled proportionally to the number of observations. If "width", all violins have the same maximum width.

size

Optional: numerical, linewith of black line around the violin plot

Value

The ggplot object of the boxplots

Details

In short, the MD-plot can be described as a PDE optimized violin plot. The Pareto Density Estimation (PDE) is an approach to estimate the probability density function (pdf) [Ultsch, 2005].

MD plot was used in [Thrun et al.,2018] for the evaluation of stochastic clustering methods and used in [Thrun et al.,2018a] in order to simultaneously estimate variances of a high-dimensional data set. The MD-plot is in the process of beeing published.

References

[Ultsch, 2005] Ultsch, A.: Pareto density estimation: A density estimation for knowledge discovery, in Baier, D.; Werrnecke, K. D., (Eds), Innovations in classification, data science, and information systems, Proc Gfkl 2003, pp 91-100, Springer, Berlin, 2005.

[Thrun et al.,2018a] Thrun, M. C., Breuer, L., & Ultsch, A. : Knowledge discovery from low-frequency stream nitrate concentrations: hydrology and biology contributions, Proc. European Conference on Data Analysis (ECDA), pp. 46-47, Paderborn, Germany, 2018.

[Thrun et al.,2018b] Thrun, M. C., Pape, F., & Ultsch, A. : Benchmarking Cluster Analysis Methods using PDE-Optimized Violin Plots, Proc. European Conference on Data Analysis (ECDA), p. 26, Paderborn, Germany, 2018.

Examples

# NOT RUN {
x <- cbind(A = runif(20000, 1, 3), B = c(rnorm(10000,0,1),rnorm(10000,2.6,1)))
MDplot(x)
# }
# NOT RUN {
#Check for significance
# }
# NOT RUN {
#requireNamespace('diptest')
#diptest::dip.test(x[,1])$p.value
#diptest::dip.test(x[,2])$p.value
# }