Analysis in R: Check data properties at once. “GGally” package

RAnalytics

This is an introduction to a package for plotting multivariate characteristics. I find this package very useful for interpreting data. It also includes a “ggmatrix” command that allows you to plot multiple ggplot objects.

Package version is 2.1.2. R version 4.2.2 is confirmed.

スポンサーリンク
Sponsored Link

Install Package

Run the following command.

#Install Package
install.packages("GGally")

Execute command

See the command and package help for details.

#Loading the library
library("GGally")

###Creating Data Examples#####
n <- 50
TestData <- data.frame(Group = sample(paste0("Group", 1:5),
                                      n, replace = TRUE),
                       Data1 = rnorm(n),
                       Data2 = rnorm(n) + rnorm(n) + rnorm(n),
                       Data3 = sample(0:1, n, replace = TRUE),
                       Data4 = sample(LETTERS[1:26], n, replace = TRUE))
#######

#Plot data features at once: ggpairs command
#Specify display columns: columns option; default:1:ncol(data)
#Specify color-coded data: mapping option
#Specify upper and lower plot contents: upper,lower option:blank to hide
##Specify content as follows,list(variable type = display contents)
##variable type;display content
##continuous; "points", "smooth", "density", "cor", "blank"
##combo; "box", "dot", "facethist", "facetdensity", "denstrip", "blank"
##discrete; "facetbar", "ratio", "blank" ##na; "na", "blank"
##Specify diagonal plot content: diag option: NULL to hide
##continuous; "densityDiag", "barDiag", "blankDiag"
##discrete; "barDiag", "blankDiag"
##na; "naDiag", "blankDiag"
##maximum number allowed in string/factor string: cardinality_threshold option
ggpairs(data = TestData, columns = c(1, 5, 2, 3),
        mapping = aes(color = Group),
        upper = list(continuous = "smooth"),
        lower = list(combo = "facetdensity"),
        diag = list(continuous = "barDiag"),
        cardinality_threshold = 30)

#Plot multiple graphs: ggmatrix command
#Specify number of rows: nrow option
#Specify number of columns: ncol option
#Specify x-axis labels: xAxisLabels option
#Specify y axis labels: yAxisLabels option
#Specify a title: title option
#Descriptive graphs are stored in a list class
PlotList <- list()
list(for (i in 1:3) {
  #BoxPlot
  PlotList[[i]] <- qplot(data = TestData, x = Group,
                         y = Data1, fill = Group, geom = "boxplot")
  #ScatterPlot
  PlotList[[i + 3]] <- qplot(data = TestData, x = Data2,
                             y = Data1, color = Group, geom = "point") +
                       ggtitle("TEST")
  })

#Plot
ggmatrix(PlotList, nrow = 2, ncol = 3,
         xAxisLabels = 1:3, yAxisLabels = 1:2, title = "TEST")

Output Examples

・ggpairs command

ggpairs

・ggmatrix command

ggmatrix

I hope this makes your analysis a little easier !!

Copied title and URL