If outliers are detected and it is deemed appropriate to exclude them, they can be removed from the data before performing standardization. If it is preferable to retain the observations, an alternate form of standardization can be conducted in which the median of the series is subtracted from each observation (as a proxy for centering) and then dividing by the interquartile range (difference between 75th and 25th percentiles) in place of the standard deviation to complete the robust scaling process.
How to perform Standardization in case of outliers?
Help us improve this post by suggesting in comments below:
– modifications to the text, and infographics
– video resources that offer clear explanations for this question
– code snippets and case studies relevant to this concept
– online blogs, and research publications that are a “must read” on this topic
Partner Ad
