Outlier detection based on extreme value theory and applications

Whether an extreme observation is an outlier or not depends strongly on the corresponding tail behavior of the underlying distribution. We develop an automatic, data‐driven method rooted in the mathematical theory of extremes to identify observations that deviate from the intermediate and central characteristics. The proposed algorithm is an extension of a method previously proposed in the literature for the specific case of heavy tailed Pareto‐type distributions to all max‐domains of attraction. We propose some applications such as a tail‐adjusted boxplot which yields a more accurate representation of possible outliers, and the identification of outliers in a multivariate context through an analysis of associated random variables such as local outlier factors. Several examples and simulation results illustrate the finite sample behavior of the algorithm and its applications.

• Publication year

  2023

• Venue

  Scandinavian Journal of Statistics

• Publication date

  2023-05-17

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1111/sjos.12665
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Benchmarking Unsupervised Outlier Detection with Realistic Synthetic Data

  2020cited by this paper
• dbscan: Fast Density-Based Clustering with R

  2019cited by this paper
• Exploratory data analysis

  2018cited by this paper
• Data-adaptive trimming of the Hill estimator and detection of outliers in the extremes of heavy-tailed data

  2018influential reference
• Estimation of the Extreme Value Index

  2016cited by this paper
• Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection

  2015cited by this paper
• Clustering of Maxima: Spatial Dependencies among Heavy Rainfall in France

  2013influential reference
• Finding a "Kneedle" in a Haystack: Detecting Knee Points in System Behavior

  2011cited by this paper
• LoOP: local outlier probabilities

  2009cited by this paper
• An adjusted boxplot for skewed distributions

  2008cited by this paper
• Extreme value theory : an introduction

  2006cited by this paper
• Ranking Outliers Using Symmetric Neighborhood Relationship

  2006cited by this paper
• Quality Assessment of Pedochemical Data Using Extreme Value Methodology

  2005cited by this paper
• Statistics of Extremes: Theory and Applications

  2005cited by this paper
• Robust and Efficient Estimation for the Generalized Pareto Distribution

  2004cited by this paper
• Enhancing Effectiveness of Outlier Detections for Low Density Patterns

  2002cited by this paper
• Robust Estimation of the Generalized Pareto Distribution

  2001cited by this paper
• Efficient algorithms for mining outliers from large data sets

  2000cited by this paper
• LOF: identifying density-based local outliers

  2000influential reference
• Robust estimation of extremes

  1998cited by this paper
• A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise

  1996cited by this paper
• Combining Instance-Based and Model-Based Learning

  1993cited by this paper
• A Simple General Approach to Inference About the Tail of a Distribution

  1975cited by this paper

• Sequential Outlier Detection in Nonstationary Time Series

  2025cites this paper
• When Tukey Meets Chauvenet: A New Boxplot Criterion for Outlier Detection

  2025cites this paper
• A new graphical device based on trimmed mean

  2025cites this paper
• ggskewboxplots: Enhanced Boxplots for Skewed Data in R

  2025cites this paper
• Unsupervised anomaly detection of industrial building energy consumption

  2024cites this paper
• Deep Learning-Based Visual Complexity Analysis of Electroencephalography Time-Frequency Images: Can It Localize the Epileptogenic Zone in the Brain?

  2023cites this paper