期刊文献+
共找到1篇文章
< 1 >
每页显示 20 50 100
Determination of Causal Effect in Observational Studies: Analysis of Correlated Data with Binary End-Points 被引量:1
1
作者 Maupi Eric Letsoalo maseka lesaoana 《Journal of Mathematics and System Science》 2012年第2期119-125,共7页
Identifying the causal impact of' some intervention challenging when one is faced with correlated binary end-points in observational studies is a challenging task, and it is even more The statistical literature on an... Identifying the causal impact of' some intervention challenging when one is faced with correlated binary end-points in observational studies is a challenging task, and it is even more The statistical literature on analyzing such data is well documented. Dependence between observations from the same study subject in correlated data renders invalid the usual chi-square tests of independence and inflates the variance ofparameter estimates. Disaggregated approaches such as hierarchical linear models which are able to adjust for individual level covariate:s are favoured in the analysis of such data, thereby gaining power over aggregated and individual-level analyses. In this article the authors, therefore, address the issue of analyzing correlated data with dichotomous end-points by using hierarchical logistic regression, a generalization of the standard logistic regression model for independent outcomes. 展开更多
关键词 Correlated data observational studies counterfactual problem hierarchical models group randomization treatment effect.
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部