Random Forest Missing Data Algorithms.
Random Forest Missing Data Algorithms.
复制标题
DOI：
10.1002/sam.11348
复制
发表时间：
2017-12
影响因子：
1.3
通讯作者：
Ishwaran H
中科院分区：
文献类型：
--
作者：
Tang F;Ishwaran H
随机森林（RF）缺失数据算法是一种有吸引力的方法填补缺失数据。它们具有能够处理混合类型的缺失数据的理想特性，它们能够适应交互和非线性，并且它们具有扩展到大数据设置的潜力。目前，有许多不同的RF插补算法，但相对较少的指导，他们的功效。使用大量不同的数据集，在不同的缺失数据机制下评估了各种RF算法的插补性能。算法包括邻近插补，在飞插补，插补利用多元无监督和监督分裂后一类代表一个新的有前途的插补算法称为missForest的推广。我们的研究结果表明，RF插补一般是强大的性能提高，增加相关性。在中度到高度缺失的情况下性能良好，甚至（在某些情况下）数据缺失不是随机的。
Random forest (RF) missing data algorithms are an attractive approach for imputing missing data. They have the desirable properties of being able to handle mixed types of missing data, they are adaptive to interactions and nonlinearity, and they have the potential to scale to big data settings. Currently there are many different RF imputation algorithms, but relatively little guidance about their efficacy. Using a large, diverse collection of data sets, imputation performance of various RF algorithms was assessed under different missing data mechanisms. Algorithms included proximity imputation, on the fly imputation, and imputation utilizing multivariate unsupervised and supervised splitting—the latter class representing a generalization of a new promising imputation algorithm called missForest. Our findings reveal RF imputation to be generally robust with performance improving with increasing correlation. Performance was good under moderate to high missingness, and even (in certain cases) when data was missing not at random.
登录
查看更多内容
影响因子：
2.3
作者：
Bartlett JW;Seaman SR;White IR;Carpenter JR;Alzheimer's Disease Neuroimaging Initiative*
通讯作者：
Alzheimer's Disease Neuroimaging Initiative*
影响因子：
7.5
作者：
Ishwaran H
通讯作者：
Ishwaran H
影响因子：
3
作者：
Liao SG;Lin Y;Kang DD;Chandra D;Bon J;Kaminski N;Sciurba FC;Tseng GC
通讯作者：
Tseng GC
影响因子：
5.8
作者：
Stekhoven, Daniel J.;Buehlmann, Peter
通讯作者：
Buehlmann, Peter
影响因子：
5
作者：
Shah, Anoop D.;Bartlett, Jonathan W.;Hemingway, Harry
通讯作者：
Hemingway, Harry