An overview of normal theory structural measurement error models

被引:7
作者
Thompson, Jeffrey R. [1 ]
Carter, Randy L.
机构
[1] N Carolina State Univ, Dept Stat, Raleigh, NC 27695 USA
[2] SUNY Buffalo, Dept Biostat, Buffalo, NY 14214 USA
关键词
instrumental variables; linear models; measurement error; nonlinear models; structural relationship;
D O I
10.1111/j.1751-5823.2007.00014.x
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
This paper gives an introduction and overview to the often under-used measurement error model. The purpose is to provide a simple summary of problems that arise from measurement error and of the solutions that have been proposed. We start by describing how measurement error models occur in real-world situations. Then we proceed with defining the measurement error model, initially introducing the multivariate form of the model, and then, starting with the simplest form of the model thoroughly discuss its features and solutions to the problems introduced due to measurement error. We discuss higher-dimensional and more advanced forms of the model and give a brief numerical illustration.
引用
收藏
页码:183 / 198
页数:16
相关论文
共 23 条