A Method of Searching for Extreme Observations in a Problem of Fuzzy Regression
Abstract
The study of statistical data for outliers is an urgent task of modern mathematics. The reliability of these methods directly affects the quality of the subsequent processing of statistical data sets and the adequacy of the resulting conclusions. In general, all available observations should be checked and compared with a certain numerical indicator. The further conclusion should be made by comparing these indicators among themselves.
In this paper, a technique to search for statistical outliers for one of the possible regression models based on the Chebyshev norm is considered. The proposed approach is based on the Legendre transformation, one of the known transformations used in the convex analysis. This algorithm allows us to refer to the group of statistical outliers for a set of observations and not for individual observations. This key point distinguishes this algorithm from the most of the commonly used algorithms. This way, the task can be solved in one pass with less required time. An example of the study of a sample for outliers is presented. The possibility to compare the obtained characteristics provides the opportunity to solve the problem for a different number of assumed extreme values.
DOI 10.14258/izvasu(2018)4-18
Downloads
Metrics
References
Tanaka H., Hayashi I., Watada J. Possibilistic Linear Regression Analysis with Fuzzy Model // European Journal of Operational Research. — 1989. — V. 40.
Дрейпер Н, Смит Г. Прикладной регрессионный анализ. Множественная регрессия = Applied Regression Analysis. — 3-е изд. — М., 2007.
Gomez A.T., Sanchez, Jorge de Andres. Applications Of Fuzzy Regression In Actuarial Analysis // Journal of Risk & Insurance. — 2003. — V. 30.
Стрижов В.В., Крымова Е.А. Методы выбора регрессионных моделей. — М., 2010.
Cook R.D. Detection of Influential Observation in Linear Regression // Technometrics. — 1977. — Vol. 19, № 1.
Andrews D.F., Pregibon D. Finding the outliers that matter // Journal of the Royal Statistical Society. — 1978. — Vol. 40.
Weisberg S. Applied linear regression, 3rd ed. — Jonh Wiley & Sans, Inc., 2005.
Пономарев И.В., Славский В.В. Нечеткая модель линейной регрессии // Доклады Академии наук. — 2009. — Т. 428, № 5.
Ponomarev I.V., Slavsky V.V. Uniformly fuzzy model of linear regression // Journal of Mathematical Sciences. — 2012. — Vol. 186, issue 3.
Куркина М.В., Пономарев И.В. Система нечетких отношений равенств в банаховом пространстве // Дифференциальные уравнения. Функциональные пространства. Теория приближений. Международная конференция, посвященная 100-летию со дня рождения С. Л. Соболева (Новосибирск, 5-12 октября 2008 г.) : тезисы докладов. — Новосибирск, 2008.
Copyright (c) 2018 И.В. Пономарев, Т.В. Саженкова, В.В. Славский
This work is licensed under a Creative Commons Attribution 4.0 International License.
Izvestiya of Altai State University is a golden publisher, as we allow self-archiving, but most importantly we are fully transparent about your rights.
Authors may present and discuss their findings ahead of publication: at biological or scientific conferences, on preprint servers, in public databases, and in blogs, wikis, tweets, and other informal communication channels.
Izvestiya of Altai State University allows authors to deposit manuscripts (currently under review or those for intended submission to Izvestiya of Altai State University) in non-commercial, pre-print servers such as ArXiv.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY 4.0) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
