On a minimum distance procedure for threshold selection in tail analysis
From MaRDI portal
Abstract: Power-law distributions have been widely observed in different areas of scientific research. Practical estimation issues include how to select a threshold above which observations follow a power-law distribution and then how to estimate the power-law tail index. A minimum distance selection procedure (MDSP) is proposed in Clauset et al. (2009) and has been widely adopted in practice, especially in the analyses of social networks. However, theoretical justifications for this selection procedure remain scant. In this paper, we study the asymptotic behavior of the selected threshold and the corresponding power-law index given by the MDSP. We find that the MDSP tends to choose too high a threshold level and leads to Hill estimates with large variances and root mean squared errors for simulated data with Pareto-like tails.
Recommendations
Cites work
- A Brief History of Generative Models for Power Law and Lognormal Distributions
- A continuous mapping theorem for the argmax‐functional in the non‐unique case
- A simple general approach to inference about the tail of a distribution
- An approximation of partial sums of independent RV'-s, and the sample DF. I
- An approximation of partial sums of independent RV's, and the sample DF. II
- An introduction to statistical modeling of extreme values
- Approximate distributions of order statistics. With applications to nonparametric statistics
- Are extreme value estimation methods useful for network data?
- Asymptotic normality of degree counts in a preferential attachment model
- Asymptotic normality of in- and out-degree counts in a preferential attachment model
- Asymptotic Statistics
- Community structure in large networks: natural cluster sizes and the absence of large well-defined clusters
- Consistency of Hill estimators in a linear preferential attachment model
- Critical points and transitions in an electric power transmission model for cascading failure blackouts
- Degree growth rates and index estimation in a directed preferential attachment model
- Directed scale-free graphs
- Emergence of Scaling in Random Networks
- Extreme value theory. An introduction.
- Fitting the linear preferential attachment model
- Goodness-of-fit tests for a heavy tailed distribution
- Heavy-Tail Phenomena
- scientific article; zbMATH DE number 3824949 (Why is no real title available?)
- Laws of large numbers for sums of extreme values
- Multivariate regular variation of discrete mass functions with applications to preferential attachment networks
- Nonstandard regular variation of in-degree and out-degree in the preferential attachment model
- On tail index estimation using dependent data
- Power-law distributions in binned empirical data
- Power-law distributions in empirical data
- Random graph dynamics
- Random graphs and complex networks. Volume 1
- Selecting the optimal sample fraction in univariate extreme value estimation
- Statistics of Extremes
- Tauberian theory for multivariate regularly varying distributions with application to preferential attachment networks
- The bootstrap methodology in statistics of extremes -- choice of optimal sample fraction
- The Pareto-Levy Law and the Distribution of Income
- Using a bootstrap method to choose the sample fraction in tail index estimation
- Vertical transmission of culture and the distribution of family names
- Weighted approximations of tail processes for \(\beta\)-mixing random variables.
Cited in
(12)- Threshold selection and trimming in extremes
- Large degrees in scale-free inhomogeneous random graphs
- Are extreme value estimation methods useful for network data?
- Consistency of Hill estimators in a linear preferential attachment model
- Gradient boosting for extreme quantile regression
- Poisson edge growth and preferential attachment networks
- Extremal properties of evolving networks: local dependence and heavy tails
- The estimation risk in extreme systemic risk forecasts
- Emergence of multivariate extremes in multilayer inhomogeneous random graphs
- Nonparametric estimation of splicing points in skewed cost distributions: a kernel-based approach
- Extremal dependence networks for mitigating systemic risk via maximum independent sets
- Randomly stopped extreme Zipf extensions
This page was built for publication: On a minimum distance procedure for threshold selection in tail analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5027018)
