Sciact
  • EN
  • RU

Number of Distinct Values in a Large Sample with Dependent Observations under fGn from an Infinite Discrete Distribution Full article

Journal Сибирские электронные математические известия (Siberian Electronic Mathematical Reports)
, E-ISSN: 1813-3304
Output data Year: 2026, Volume: 23, Number: 1, Pages: 393–403 Pages count : 11 DOI: 10.33048/semi.2026.23.025
Tags urn scheme, fractional noise, transform of Gaussian sequence, long-range dependence, statistical text modeling.
Authors Arkashov N.S. 1
Affiliations
1 Sobolev Institute of Mathematics

Funding (1)

1 Sobolev Institute of Mathematics FWNF-2024-0001

Abstract: The growth dynamics of the number of distinct values in samples obtained from a stationary sequence of dependent observations with an infinite discrete distribution is investigated. The analysis of this behavior for samples formed from a sequence of i.i.d. random variables is well-established. In this paper, the expected number of distinct values in the independent case is compared with that for dependent observations. A connection is established between the estimation of these expectations and the problem of estimating multivariate normal distributions. The application of the considered stationary sequences to statistical text modeling is discussed.
Cite: Arkashov N.S.
Number of Distinct Values in a Large Sample with Dependent Observations under fGn from an Infinite Discrete Distribution
Сибирские электронные математические известия (Siberian Electronic Mathematical Reports). 2026. V.23. N1. P.393–403. DOI: 10.33048/semi.2026.23.025
Dates:
Submitted: Dec 11, 2025
Published online: Apr 20, 2026
Identifiers: No identifiers
Altmetrics: