Prediction with missing data via Bayesian Additive Regression Trees
Prediction with missing data via Bayesian Additive Regression Trees
复制标题
DOI：
10.1002/cjs.11248
复制
发表时间：
2015-06-01
影响因子：
0.6
通讯作者：
Bleich, Justin
中科院分区：
文献类型：
--
作者：
Kapelner, Adam;Bleich, Justin
我们提出了一种利用非参数统计学习将缺失数据合并到一般预测问题中的方法。我们重点研究了一种基于树的方法，贝叶斯加性回归树(BART)，它是最近提出的一种将遗漏纳入决策树的方法，它在属性中加入了遗漏。此过程扩展了在基于树的模型中找到的本机分区机制，并且不需要归罪。对生成的模型和真实数据的模拟表明，我们的过程为选择模型和模式混合框架提供了希望，这是通过样本外预测精度来衡量的。我们还说明了BART将失配纳入不确定区间的能力。我们的实现在R包bartMachine中随时可用。《加拿大统计杂志》43：224-239；2015(C)2015年加拿大统计学会
We present a method for incorporating missing data into general prediction problems which use nonparametric statistical learning. We focus on a tree-based method, Bayesian Additive Regression Trees (BART), enhanced with Missingness Incorporated in Attributes, a recently proposed approach for incorporating missingness into decision trees. This procedure extends the native partitioning mechanisms found in tree-based models and does not require imputation. Simulations on generated models and real data indicate that our procedure offers promise for both selection model and pattern-mixture frameworks as measured by out-of-sample predictive accuracy. We also illustrate BART's abilities to incorporate missingness into uncertainty intervals. Our implementation is readily available in the R package bartMachine. The Canadian Journal of Statistics 43: 224-239; 2015 (c) 2015 Statistical Society of Canada