ABSTRACT
After plotting a frequency histogram, an event that is much larger or much smaller than the remainder of the sample may be evident. This will initially be labeled as an
extreme event
. Some data samples may contain more than one extreme event. Extreme events can create problems in data analysis and modeling. For example, an extremely large value can cause the sample mean and standard deviation to be much larger than the population values. In bivariate analysis (i.e.,
X
vs.
