Exponential convergence rates for the kernel bivariate distribution function estimator under NSD assumption with application to hydrology data. Issue 12 (20th May 2022)

• Record Type: Journal Article
• Authors: Kheyri, A.; Amini, M.; Jabbari, H.; Bozorgnia, A.; Volodin, A.
• Is Part Of: Communications in statistics. Volume 51:Issue 12(2022)
• Page Start: 4042
• Page End: 4054
• Publication Date: 2022-05-20
• Subjects: Bivariate distribution function -- kernel estimation -- negative superadditive dependence; Primary 62G07 -- Secondary 62G20; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2020.1808900
• Languages: English
• ISSNs: 0361-0926

Abstract

Abstract: In this paper, we study the asymptotic behavior of the kernel bivariate distribution function estimator for negatively superadditive dependent. The exponential convergence rates for the kernel estimator are investigated. Under certain regularity conditions, the optimal bandwidth rate is determined with respect to mean squared error criteria. A simulation study is used to justify the behavior of the kernel and histogram estimators. As an application, a real data set in hydrology is considered and the kernel bivariate distribution function estimator of the data is investigated.