Data Visualization in R and Python

12Marginals and Plots Alignment

So-called marginals are a family of graphics made by the combination of different plots with a main one presented in the central position and one or two others associated to the x and y axes. For example, we may have a scatterplot as the main graphic and histograms, density plots, or boxplots associated to the axes. Several other variants are possible.

Dataset

In this chapter, we make use again of data from Bicycle thefts in Berlin (transl. Fahrraddiebstahl in Berlin) from the Municipality of Berlin, Germany, Berlin Open Data, previously introduced.

12.1 R: ggplot

Dataset read and change of column names are the same as already shown before, here omitted. We aggregate values for year, month, and bike type, calculating bike values, and number of stolen bikes.

bikesR= group_by(df,
               year(DATE),
               month(DATE, label=TRUE, abbr=FALSE),
               TYPE_OF_BICYCLE) %>%
      summarize(TOT_DMG= sum(DAMAGES), NUM =n()) %>%
      rename(YEAR = 1, MONTH_CREATED = 2)
 

12.1.1 Marginal

Features to produce marginal graphics are offered by package ggExtra with function ggMarginal().

We create a first simple example by defining a scatterplot with some style elements and assigning it to variable p. For convenience, we omit the two bike types with the lowest number of thefts and plot the number of thefts with respect to the average value of stolen bikes for bike type. We add also shape as an aesthetic to observe the effect on black and white support, as is the paper edition of the ...