ABSTRACT
A very important type of test that estimators and other analysts should perform, but perhaps one that is not always performed quite as formally as it might is in the detection of outliers. An outlier is a value that falls substantially outside the pattern of other data. The outlier may be representative of unintended atypical factors or may simply be a value which has a very low probability of occurrence. The Inner Tukey Fences sound like a reasonable basis for identifying potential outliers. Chauvenet's Criterion is based on that assumption of Normality. It calculates the number of data points estimators might reasonably expect to get from a known sample size based on the Cumulative Distribution Function (CDF) of a Normal Distribution; in effect it uses the Z-Score. Frank Grubbs proposed a test to detect a SINGLE outlier. It compares the largest Absolute Deviation from the Mean of the data points in a sample to their Standard Deviation.
