Quantile regression for longitudinal data
Quantile regression for longitudinal data
复制标题
DOI：
10.1016/j.jmva.2004.05.006
复制
发表时间：
2004-10-01
影响因子：
1.6
通讯作者：
Koenker, R
中科院分区：
文献类型：
--
作者：
Koenker, R
经典随机效应估计量的惩罚最小二乘解释为具有大量“固定效应”的分位数回归模型提供了一种可能的方法。大量个体固定效应的引入会显著夸大其他协变量效应估计的可变性。正则化，或将这些个体效应缩小到一个公共值，可以帮助修正这种通胀效应。利用L(1)的正则化方法，提出了估计纵向数据分位数回归模型的一般方法。稀疏线性代数和求解大型线性规划的内点方法是重要的计算工具。(C)2004 Elsevier Inc.保留所有权利。
The penalized least squares interpretation of the classical random effects estimator suggests a possible way forward for quantile regression models with a large number of "fixed effects". The introduction of a large number of individual fixed effects can significantly inflate the variability of estimates of other covariate effects. Regularization, or shrinkage of these individual effects toward a common value can help to modify this inflation effect. A general approach to estimating quantile regression models for longitudinal data is proposed employing l(1) regularization methods. Sparse linear algebra and interior point methods for solving large linear programs are essential computational tools. (C) 2004 Elsevier Inc. All rights reserved.