Nonparametric algorithms for recovery of mutually unbeatted functions on observations


Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций

Конференция: International Workshop Applied Methods of Statistical Analysis Nonparametric Methods in Cybernetics and System Analysis, AMSA 2017; Красноярск, Россия; Красноярск, Россия

Год издания: 2017

Ключевые слова: A nonparametric model, A priori information, Mutually ambiguous characteristics, Nonparametric estimates

Аннотация: The problem of reconstructing a function from observations with random errors is considered. Moreover, at the formulation stage of the problem there is no stage associated with the parametric structure of this function. Therefore, the estimate is sought in the class of nonparametric statistics, when the original description of the Показать полностьюfunction is unknown up to a parameter vector. The peculiarity of this problem is that the desired function is described by a mutually ambiguous characteristic and the generally accepted nonparametric estimation proves to be unsuitable. It was necessary to introduce a new class of nonparametric estimators. The results of some computational experiments are presented. © Novosibirsk State Technical University, 2017.

Ссылки на полный текст


Журнал: Applied Methods of Statistical Analysis

Номера страниц: 64-71

ISSN журнала: 2313870X

Издатель: Novosibirsk State Technical University


  • Korneeva A.A. (Siberian Federal University, Krasnoyarsk, Russian Federation)
  • Chernova S.S. (Siberian Federal University, Krasnoyarsk, Russian Federation)
  • Shishkina A.V. (Siberian Federal University, Krasnoyarsk, Russian Federation)

Вхождение в базы данных