ImpactEffectsize (version 0.3.0)

Impact: Impact effect size measure

Description

Calculates the Impact effect size measure that is based on the group distance and the difference in the shape of the data distribution between two groups.

Usage

Impact(Data, Cls, PlotIt = FALSE, pde = TRUE, 
  col = c("red","blue"), meanLines = FALSE, medianLines = FALSE, ...)

Arguments

Data

the data of both groups as a vector.

Cls

the class information as a vector of similar length as the data vector.

PlotIt

plots the pdf of the two groups as Pareto density estimation.

col

colors of the two lines to plot. Will be ignored if PlotIt = FALSE.

pde

plots a standard pdf insted of a non-parametric Pareto density estimation. Will be ignored if PlotIt = FALSE.

meanLines

adds perpendicular lines at the means of the two groups. Will be ignored if PlotIt = FALSE.

medianLines

adds perpendicular lines at the medians of the two groups. Will be ignored if PlotIt = FALSE.

…

more graphical parameters can be given as plot arguments if PlotIt = TRUE.

Value

Returns a list of value incuding the main result and intermediate results.

Impact

the main effect size measure.

MorphDiff

the extend of the group-difference in the shapes of the pdf.

CTDiff

the extend of the difference in the group medians.

GMDdata

Gini's mean difference as a measure of data variabilty pooled between groups.

References

Lotsch, J., and Ultsch, A. (2019): ImpactEffectsize -- an R Package for calculation and visualisation of the Impact distance and shape based effect size measure.

Examples

# NOT RUN {
## example 1
data("FeatureselectionData")
ImpactSize <- Impact(FeatureselectionData$Var0011, FeatureselectionData$Classes)$Impact
plot2Densities(Data = FeatureselectionData$Var0011, Cls = FeatureselectionData$Classes, 
               xlab = "Variable", ylab = "Density", main = paste0("Impact = ",round(ImpactSize,3))) 
# }