Chernoff's faces: A method or technique for representing the multivariate data graphically. Each observation is represented by the computer-created face, the features of which are controlled by an observation's variable values. The collection of the faces representing the set of observations might be useful in identifying the groups of similar outliers, individuals, etc are shown in the figure 30.
and where with x1; y1 ... xn; yn being observed sample values, and the I(A) being the indicator function of the event A. Example plots are shown in the figure 31. Part (a) shows the situation in which the x and y are independent, Part (b) in which they have the correlation of 0.6. In each and every case the left hand plot is a simple scatter plot of the data, and the right hand is consequent chi-plot.