The first-order random coefficient integer valued autoregressive process with the occasional level shift random noise based on dual empirical likelihood. Issue 12 (17th June 2020)

• Record Type: Journal Article
• Authors: Zhang, Shuxia; Cong, Xinrong; Tian, Boping; Li, Yanpeng; Yao, Mingjun
• Is Part Of: Communications in statistics. Volume 49:Issue 12(2020)
• Page Start: 2990
• Page End: 3009
• Publication Date: 2020-06-17
• Subjects: Empirical likelihood -- random coefficient -- semi-parametric -- asymptotic distribution -- logistic regression -- limit theory; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2019.1584315
• Languages: English
• ISSNs: 0361-0926

Abstract

Abstract: This paper investigates the first-order random coefficient integer valued autoregressive process with the occasional level shift random noise based on dual empirical likelihood. The limiting distribution of log empirical likelihood ratio statistic is constructed. Asymptotic convergence and confidence region results of empirical likelihood ratio are given. Hypothesis testing is considering, and maximum empirical likelihood estimation for parameter is acquired. Simulations are given to show that the maximum empirical likelihood estimation is more efficient than the conditional least squares estimation.