Robust replicated heteroscedastic measurement error model using heavy-tailed distribution. Issue 6 (3rd July 2018)

• Record Type: Journal Article
• Authors: Cao, Chunzheng; Chen, Mengqian; Ren, Yuqian; Xu, Yue
• Is Part Of: Communications in statistics. Volume 47:Issue 6(2018)
• Page Start: 1771
• Page End: 1784
• Publication Date: 2018-07-03
• Subjects: EM algorithm -- Heteroscedasticity -- Replicated measurement -- Robustness -- Scale mixtures of normal distributions; 62J05 -- 62F35; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2017.1324979
• Languages: English
• ISSNs: 0361-0918

Abstract

ABSTRACT: Heteroscedastic measurement error models are widely used in epidemiological, analytical chemistry, and other research areas. In this article, we propose a heteroscedastic measurement error model for replicated data under scale mixtures of normal distributions with/without equation error, which covers unpair and/or unequal replication cases. We obtain iterative formulas of maximum likelihood estimations via EM algorithm, and provide closed forms of asymptotic variances of the estimators. Simulation studies and a real data application are reported to investigate the effective and robust performances of the model and estimates.