Tutorial: Neutralizing Outliers in Any Dimension

In this article, we discuss a general framework to drastically reduce the influence of outliers in most contexts. It applies to problems such as clustering (finding centroids,) regression, measuring correlation or R-Squared, and many more. We will focus on the centroid problem here, as it is very similar and generalizes easily to solving a linear regression. The correlation / R-Squared issue was discussed in an earlier article and involves only a change of formula. Clustering and regression are more complex problems involving iterative algorithms.