In Chapter 3, Data Visualization, we saw the effectiveness of boxplot. For independent variables, we can simply draw separate boxplots for the variables and visualize the distribution. However, when there is dependency between two variables, distinct boxplot loses the dependency among the two variables. Thus, we need to see if there is a way to visualize the data through a boxplot. The answer to the question is provided by bagplot or bivariate boxplot.
The bagplot characteristic is described in the following steps:
*in the bagplot, is the point with the highest half space depth.