Multiple imputation of incomplete zero‐inflated count data

Multiple imputation of incomplete zero‐inflated count data
复制标题

不完整的零膨胀计数数据的多重插补

DOI:
10.1111/stan.12009
复制
发表时间:
2013
影响因子:
1.5
通讯作者:
Reinecke
Reinecke
中科院分区:
数学4区
文献类型:
--
作者:
Kleinke;Reinecke

文献摘要

参考文献

被引文献

相似文献

经验计数数据通常是零膨胀且过度分散的。目前,没有软件包可以对这些数据进行充分的插补。我们基于贝叶斯回归方法或基于引导方法提出了针对此类计数数据的多重插补例程,该方法作为 R 中链式方程(小鼠)软件流行的多重插补的附加程序(van BuurenandGroothuis-Oudshoorn,统计软件杂志,第 45 卷,2011 年,第 1 页)。我们在蒙特卡罗模拟中证明,我们的程序优于当前可用的计数数据程序。需要强调的是,彻底的建模对于获得合理的插补至关重要,并且模型的错误指定可能会使参数估计和标准误差产生相当明显的偏差。最后,讨论了我们程序的优点和局限性,并概述了未来理论和软件开发的富有成效的途径。
Empirical count data are often zero‐inflated and overdispersed. Currently, there is no software package that allows adequate imputation of these data. We present multiple‐imputation routines for these kinds of count data based on a Bayesian regression approach or alternatively based on a bootstrap approach that work as add‐ons for the popular multiple imputation by chained equations (mice) software in R (van BuurenandGroothuis‐Oudshoorn,Journal of Statistical Software, vol. 45, 2011, p. 1). We demonstrate in a Monte Carlo simulation that our procedures are superior to currently available count data procedures. It is emphasized that thorough modeling is essential to obtain plausible imputations and that model mis‐specifications can bias parameter estimates and standard errors quite noticeably. Finally, the strengths and limitations of our procedures are discussed, and fruitful avenues for future theory and software development are outlined.
DOI: 10.1016/0167-9473(95)00057-7
发表时间: 1996-08-10
影响因子: 1.8
作者:
Schenker, N;Taylor, JMG
通讯作者: Taylor, JMG
DOI: 10.1037/1082-989x.6.4.330
发表时间: 2001-12
影响因子: 7
作者:
L. Collins;J. Schafer;Chi-Ming Kam
通讯作者: L. Collins;J. Schafer;Chi-Ming Kam
DOI: 10.1177/0962280206074464
发表时间: 2007-01-01
影响因子: 2.3
作者:
Yu, L-M;Burton, Andrea;Rivero-Arias, Oliver
通讯作者: Rivero-Arias, Oliver
“R 中的心理测量学”特卷简介
DOI: --
发表时间: 2007
期刊:
影响因子: --
作者:
J. Leeuw;P. Mair
通讯作者: P. Mair
DOI: 10.1038/nprot.2008.110
发表时间: 2008-01-01
期刊: NATURE PROTOCOLS
影响因子: 14.8
作者:
Woo, Hin-Koon;Northen, Trent R.;Siuzdak, Gary
通讯作者: Siuzdak, Gary