Regression Analysis — Outliers and other Influential Observations
Exhibit 33.30 Box and Whiskers plot. This indicates whether distribution is skewed and reveals outliers, i.e., values lying beyond the whiskers (data point #5).
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 33.30, 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.
Previous Next
Use the Search Bar to find content on MarketingMind.
Marketing Analytics Workshop
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.
Digital Marketing Workshop
Unlock the Power of Digital Marketing: Join us for an immersive online experience designed to empower you with the skills and knowledge needed to excel in the dynamic world of digital marketing. In just three days, you will transform into a proficient digital marketer, equipped to craft and implement successful online strategies.
