Heterogeneous treatment effect analysis based on machine-learning methodology.

Heterogeneous treatment effect analysis based on machine-learning methodology.
复制标题

DOI:
10.1002/psp4.12715
复制
发表时间:
2021-11
期刊:
CPT: pharmacometrics & systems pharmacology
影响因子:
--
通讯作者:
Zhao L
Zhao L
中科院分区:
其他
文献类型:
--
作者:
Gong X;Hu M;Basu M;Zhao L

文献摘要

参考文献

被引文献

相似文献

异质治疗效应(HTE)分析侧重于检查群体中个体或亚组的不同治疗效果。例如,知情的理解可以关键地指导医生对某种疾病进行个性化的医疗治疗。然而,即使大数据时代的到来带来了数据可用性的爆炸性增长,hte分析也没有得到广泛的认可和使用。其使用不足的部分原因是,数据往往是高维和高复杂性的,这对应用传统的HTE分析方法构成了巨大的挑战。为了应对这些挑战,在随机森林机器学习算法的基础上提出了一种新的因果森林HTE方法。我们通过模拟不同复杂程度的场景对因果森林方法和传统的两步法进行了系统的性能评估。我们的结果表明,因果森林在评估治疗效果方面优于传统的HTE方法,特别是在数据复杂(例如,非线性)和高维的情况下,这表明因果森林是一种很有前途的HTE分析的现实应用工具。
Heterogeneous treatment effect (HTE) analysis focuses on examining varying treatment effects for individuals or subgroups in a population. For example, an HTE‐informed understanding can critically guide physicians to individualize the medical treatment for a certain disease. However, HTE analysis has not been widely recognized and used, even given the explosive increase of data availability attributed to the arrival of the Big Data era. Part of the reason behind its underuse is that data are often of high dimension and high complexity, which pose significant challenges for applying conventional HTE analysis methods. To meet these challenges, a newly developed causal forest HTE method has been derived from the random forest machine‐learning algorithm. We conducted a systematic performance evaluation for the causal forest method against the conventional two‐step method by simulating scenarios with different levels of complexity for the analysis. Our results show that causal forest outperforms the conventional HTE method in assessing treatment effect, especially when data are complex (e.g., nonlinear) and high dimensional, suggesting that causal forest is a promising tool for real‐world applications of HTE analysis.
DOI: 10.1214/18-aos1709
发表时间: 2019-04-01
影响因子: 4.5
作者:
Athey, Susan;Tibshirani, Julie;Wager, Stefan
通讯作者: Wager, Stefan
DOI: 10.1093/poq/nfs036
发表时间: 2012-09-01
影响因子: 3.4
作者:
Green, Donald P.;Kern, Holger L.
通讯作者: Kern, Holger L.
DOI: 10.1002/jae.814
发表时间: 2005-12-01
影响因子: 2.1
作者:
Fernández-Villaverde, J;Rubio-Ramírez, JF
通讯作者: Rubio-Ramírez, JF
DOI: 10.1038/nbt0908-1011
发表时间: 2008-09
影响因子: 46.9
作者:
Kingsford, Carl;Salzberg, Steven L.
通讯作者: Salzberg, Steven L.
DOI: 10.1073/pnas.1510489113
发表时间: 2016-07-05
影响因子: 11.1
作者:
Athey, Susan;Imbens, Guido
通讯作者: Imbens, Guido