Quantifying causality in data science with quasi-experiments.

Quantifying causality in data science with quasi-experiments.
复制标题

DOI:
10.1038/s43588-020-00005-8
复制
发表时间:
2021-01
期刊:
NATURE COMPUTATIONAL SCIENCE
影响因子:
--
通讯作者:
Kording, Konrad
Kording, Konrad
中科院分区:
其他
文献类型:
--
作者:
Liu, Tony;Ungar, Lyle;Kording, Konrad

文献摘要

参考文献

被引文献

相似文献

从观测数据中估计因果关系在许多数据科学问题中至关重要,但可能是一项具有挑战性的任务。在这里,我们回顾了计量经济学中流行的因果关系方法,这些方法利用了现有数据中的(准)随机变化,称为准实验,并展示了它们如何与机器学习相结合,以回答典型数据科学环境中的因果问题。我们还强调了数据科学家如何帮助推进这些方法,为医学、工业和社会的高维数据带来因果估计。
Estimating causality from observational data is essential in many data science questions but can be a challenging task. Here we review approaches to causality that are popular in econometrics and that exploit (quasi) random variation in existing data, called quasi-experiments, and show how they can be combined with machine learning to answer causal questions within typical data science settings. We also highlight how data scientists can help advance these methods to bring causal estimation to high-dimensional data from medicine, industry and society.
DOI: 10.3982/ecta9626
发表时间: 2012-11-01
期刊: ECONOMETRICA
影响因子: 6.1
作者:
Belloni, A.;Chen, D.;Hansen, C.
通讯作者: Hansen, C.
DOI: 10.1109/access.2018.2870052
发表时间: 2018-01-01
期刊: IEEE ACCESS
影响因子: 3.9
作者:
Adadi, Amina;Berrada, Mohammed
通讯作者: Berrada, Mohammed
DOI: 10.1257/jep.15.4.69
发表时间: 2001-09-01
影响因子: 8.4
作者:
Angrist, JD;Krueger, AB
通讯作者: Krueger, AB
DOI: 10.1111/1468-0297.00578
发表时间: 2000-11-01
期刊: ECONOMIC JOURNAL
影响因子: 3.2
作者:
Besley, T;Case, A
通讯作者: Case, A
DOI: 10.1146/annurev-economics-080217-053402
发表时间: 2018-01-01
期刊: ANNUAL REVIEW OF ECONOMICS, VOL 10
影响因子: --
作者:
Abadie, Alberto;Cattaneo, Matias D.
通讯作者: Cattaneo, Matias D.