An observation in a data set that lies within the interior of a distribution but is in error. For example, the observation may be expressed in euros whereas all other observations are measured in US dollars. The outliers in a data set are obvious. In contrast, the inliers are difficult to detect.