Robust statistics for outlier detection

TLDR

Outlying observations can strongly influence results, causing problems in data analysis. The study reviews robust methods and tools for detecting outliers by fitting models to the majority of data. The authors discuss robust procedures for univariate, low‑dimensional, and high‑dimensional data, including location and scatter estimation, linear regression, PCA, and classification. © 2011 John Wiley & Sons, Inc.; DOI: 10.1002/widm.2.

Abstract

Abstract When analyzing data, outlying observations cause problems because they may strongly influence the result. Robust statistics aims at detecting the outliers by searching for the model fitted by the majority of the data. We present an overview of several robust methods and outlier detection tools. We discuss robust procedures for univariate, low‐dimensional, and high‐dimensional data such as estimation of location and scatter, linear regression, principal component analysis, and classification. © 2011 John Wiley & Sons, Inc. WIREs Data Mining Knowl Discov 2011 1 73‐79 DOI: 10.1002/widm.2 This article is categorized under: Algorithmic Development > Biological Data Mining Algorithmic Development > Spatial and Temporal Data Mining Application Areas > Health Care Technologies > Structure Discovery and Clustering