Databases Reference
In-Depth Information
Figure 3.5
Scatter Plot for Table6.csv
A cursory review of the plot does not reveal any obvious outliers.
Rotate the plot to reveal its back side by dragging horizontally. (See
Figure 3.6.)
Do you see the outlier? Although not as preattentive as in the scatter plot, the
outlier is visible. It is represented by the connecting line between var1 and var2
that runs in the opposite direction from the other observations. (Figure 3.6.)
If you have trouble locating the outlier, it may help to watch as you rotate the
plot slightly.
Note that in the parallel plot, pattern-based outliers are only visible between
adjacent columns. To view all possible combinations, systematically reorder the
columns. A systematic search is more effective using the correlation matrix/-
scatter plot combination rather than the parallel plot. We present the parallel
plot view of outliers for two reasons. First, as you are using the parallel plot for
other exploration purposes, you may run across pattern-based outliers and
should be able to recognize them. Second, the parallel plot is effective in
isolating and removing the outlier.
Drag the top var2 range slider down, approaching the outlying observation.
Getting really close is not necessary; just get close enough to see the outlier
clearly isolated on the var1 axis.
