文章基本信息

标题：A Weighted Composite Likelihood Approach to Inference from Clustered Survey Data Under a Two-Level Model
本地全文：下载
作者：Laura Dumitrescu ; Wei Qian ; J. N. K. Rao 等
期刊名称：Sankhya. Series A, mathematical statistics and probability
印刷版ISSN：0976-836X
电子版ISSN：0976-8378
出版年度：2021
卷号：83
期号：2
页码：814-843
DOI：10.1007/s13171-020-00234-z
语种：English
出版社：Indian Statistical Institute
摘要：Two-level models are widely used for analysing clustered survey data with the design structure matching the model hierarchy. Hypothesis testing on model parameters is studied, using a weighted composite likelihood approach that takes account of the survey design features. In particular, the asymptotic normality of the weighted composite likelihood estimators is established. Using this result, the asymptotic distributions of a generalised score test statistic and a likelihood ratio type test statistic, under a null composite hypothesis on the model parameters, is established. Results of a limited simulation study on the finite sample performance of the proposed tests are reported.