An outlier is an observation that is distant from other observations. It may result from measurement error; in which case it should be discarded. Or it may be indicative of a heavy-tailed population distribution, which then violates the assumption of normality.
Box and whisker plots such as the one shown in Exhibit 33.29, reveal outliers in a univariate assessment. For pairs of variables, outliers appear as isolated points on the outskirts of scatterplots. For more than 2 variables, statistical techniques such as Mahalanobis D2 may be used for detecting outliers.
Influential observations are any observations, outliers included, that have a disproportionate effect on the regression results. These need to be carefully examined and should be removed, unless there is a rationale for retaining them.
Use the Search Bar to find content on MarketingMind.
In an analytics-driven business environment, this analytics-centred consumer marketing workshop is tailored to the needs of consumer analysts, marketing researchers, brand managers, category managers and seasoned marketing and retailing professionals.
Is marketing education fluffy too?
Marketing simulators impart much needed combat experiences, equipping practitioners with the skills to succeed in the consumer market battleground. They combine theory with practice, linking the classroom with the consumer marketplace.