Local polynomial regression analysis of clustered data

• 成果类型: Article
• 署名作者: Chen, KN; Jin, ZZ
• 署名单位: Hong Kong University of Science & Technology; Columbia University
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/92.1.59
• 发表日期: 2005
• 页码: 5974
• 关键词: nonparametric function estimation; generalized linear-models; Semiparametric models; variable bandwidth; kernel regression; longitudinal data; smoothers

摘要

This paper proposes a classical weighted least squares type of local polynomial smoothing for the analysis of clustered data, with the key idea of using generalised inverses of correlation matrices. The estimator has a simple closed-form expression. Simplicity is achieved also for nonparametric generalised linear models with arbitrary link function via a transformation. Our approach can be characterised by 'local observations with local variances', which yields intuitively correct results in the sense that correct/incorrect specification of within-cluster correlation has respective positive/negative effects. The approach is a natural extension of classical local polynomial smoothing. Consequently, existing theory can be largely carried over and important issues such as bandwidth selection can be tackled in the classical fashion. Moreover, the approach can handle various types of covariate, such as cluster-level, subject-level or partially cluster-level. Numerical studies support the theoretical results. The method is illustrated with a real example on luteinising hormone levels in cows.