首页> 外文期刊>Journal of applied statistics >Generalized Linear Models with Random Effects: Unified Analysis via H-likelihood
【24h】

Generalized Linear Models with Random Effects: Unified Analysis via H-likelihood

机译:具有随机效应的广义线性模型:通过H可能性进行统一分析

获取原文
获取原文并翻译 | 示例
       

摘要

This book, with its 12 chapters, introduces statistical modelling and inference on different types of linear models with random effects using various likelihoods. Chapter 1 gives a survey of likelihood theory starting from estimation of all fixed parameters to a subset of them via (modified) profile likelihood, including various types of likelihoods, random parameters and Bayesian approaches. Chapter 2 begins with building linear models via various types of linear predictor including the aliasing effect and their generalizations (GLMs). The goodness of fit, discrepancy and parameter estimation of these models are verified. Chapter 3 proposes quasi-likelihood (QL) and its properties, including QL estimators of regression parameters via the iteratively weighted least squares method, Hessian matrices and sandwich formulae.
机译:本书共十二章,介绍了统计建模和对具有不同可能性的随机效应的不同类型线性模型的推断。第1章对似然理论进行了概述,从估计所有固定参数到通过(修改的)轮廓似然(包括各种类型的似然,随机参数和贝叶斯方法)对其子集进行了估计。第2章从通过各种类型的线性预测变量建立线性模型开始,包括混叠效应及其概括(GLM)。验证了这些模型的拟合优度,差异和参数估计的优劣。第3章提出了拟似然(QL)及其性质,包括通过迭代加权最小二乘法,Hessian矩阵和三明治公式对回归参数进行QL估计。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号