Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/39367
Title: | A new distance based measure of asymmetry | Authors: | BAILLIEN, Jonas Gijbels, Irene VERHASSELT, Anneleen |
Issue Date: | 2023 | Publisher: | ELSEVIER INC | Source: | JOURNAL OF MULTIVARIATE ANALYSIS, 193 (Art N° 105118) | Abstract: | In a univariate setting there is a near unanimous agreement on the notion of skewness. Nevertheless, many more skewness measures, or also called asymmetry measures (or indices) exist, each with their benefits. Extending the concept of skewness or asymmetry to a multivariate setting is a much harder problem. Attempts have been made, but the unanimity of the univariate setting is no longer present. Most asymmetry indices are scalar or vector based measures, but this can lead to a loss of information concerning asymmetry. To this end, we propose a novel functional asymmetry index which is based on the natural idea of reflective symmetry around the mode. The proposed index is also extended to the multivariate setting and a summarizing scalar (or vector based index in multivariate setting) is derived from it.(c) 2022 Elsevier Inc. All rights reserved. | Notes: | Gijbels, I (corresponding author), Katholieke Univ Leuven, Dept Math, Celestijnenlaan 200B,Box 2400, B-3001 Leuven, Heverlee, Belgium. irene.gijbels@kuleuven.be |
Keywords: | Asymmetry measure;Mode;Nonparametric estimation;Parametric estimation;Reflective symmetry;Uniform consistency | Document URI: | http://hdl.handle.net/1942/39367 | ISSN: | 0047-259X | DOI: | 10.1016/j.jmva.2022.105118 | ISI #: | 000909030500013 | Rights: | 2022 Elsevier Inc. All rights reserved. | Category: | A1 | Type: | Journal Contribution |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
main.pdf | Published version | 2.19 MB | Adobe PDF | View/Open |
A new distance based measure of asymmetry_.pdf Until 2025-01-31 | Peer-reviewed author version | 1.78 MB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.