Model Averaging for Linear Mixed-Effects Models May 11, 2012

摘要：

We study model averaging in linear mixed-effects models. When the covariance matrix of random effects is known, the weight vector is obtained by minimizing the well-known Mallows criterion. When the covariance matrix is unknown, we develop an unbiased estimator of squared risk of model averaging, and use the estimator as a new criterion for weight choice. The corresponding model average estimator is shown to be asymptotically optimal. Simulation experiments and application in AIDS clinic trails have provided evidence of the superiority of the proposed model averaging method.

关键词： Asymptotic optimality Mixed-effects Model Model averaging Conditional Akaike information criterion

作者: Xinyu Zhang

作者单位: Academy of Mathematics and Systems Science,Chinese Academy of Sciences,Beijing 100190,P. R.China Center of Forecasting Science,Chinese Academy of Sciences,Beijing 100190,P. R. China

会议类型: 国际会议

会议名称: Second Joint Biostatistics Symposium(第二届生物统计国际研讨会2012)

会议地点: 北京

会议语种:英文

页码: 384-405

在线出版日期: 2012-07-08（万方平台首次上网日期，不代表论文的发表时间）