What is Outlier?

An outlier is a value that is far from the rest of the data, which may be a genuine extreme or an error.

Outlier explained

Outliers matter because they distort averages and totals. Before removing one, find out what it is: a typo (an order of 10,000 instead of 100), a test record, or a real but unusual customer.

Common ways to flag them are a z-score beyond about 3 standard deviations, or a value more than 1.5 times the interquartile range outside the quartiles. These are conventions, not laws.

Example

One order for 250,000 among thousands of orders under 500 is an outlier that could be a bulk purchase or a data-entry error.

Frequently asked questions

What is Outlier?
An outlier is a value that is far from the rest of the data, which may be a genuine extreme or an error.
Can you give an example of Outlier?
One order for 250,000 among thousands of orders under 500 is an outlier that could be a bulk purchase or a data-entry error.