Introducing box plots
A box plot shows the distribution of a numeric column without assuming any parametric distribution of the data. A parametric method would be—for example—the z-score method, which calculates how far in the tails of a Gaussian distribution a data point appears. A box plot instead reports sample quantiles of the data and shows the variability of the data relative to the range between the first and third quartiles. It is also possible to report variability by groups using a conditional box plot.
In the following subsections, we will explain how you can use a box plot to visually explore time series data and how you can build a (conditional) box plot in KNIME.
Inspecting variability of data in a box plot
A box plot shows the first quartile (Q1), median (Q2), and third quartile (Q3) that make the box-plot body, and in addition, the whiskers that determine the range of normal variation. The upper whisker is calculated by adding to Q3 the length...
