Mixture Models: Latent Profile and Latent Class Analysis
Mixture Models: Latent Profile and Latent Class Analysis
复制标题
DOI：
10.1007/978-3-319-26633-6_12
复制
发表时间：
2016-01-01
期刊：
影响因子：
--
通讯作者：
Oberski, Daniel
中科院分区：
文献类型：
--
作者：
Oberski, Daniel
旨在从观察到的数据中恢复隐藏组。它们类似于聚类技术，但更灵活，因为它们基于数据的显式模型，并允许您考虑恢复的组是不确定的这一事实。当您想要将大量连续（LPA）或分类（LCA）变量减少到几个子组时，LCA和LPA非常有用。它们还可以帮助实验者在不同的人的治疗效果不同的情况下，但我们不知道哪些人。本章解释了LPA和LCA是如何工作的，这些技术背后的假设是什么，以及如何使用R来应用它们。
that aim to recover hidden groups from observed data. They are similar to clustering techniques but more flexible because they are based on an explicit model of the data, and allow you to account for the fact that the recovered groups are uncertain. LCA and LPA are useful when you want to reduce a large number of continuous (LPA) or categorical (LCA) variables to a few subgroups. They can also help experimenters in situations where the treatment effect is different for different people, but we do not know which people. This chapter explains how LPA and LCA work, what assumptions are behind the techniques, and how you can use R to apply them.