/awesome-causality-data

A data index for learning causality.

Primary LanguageJupyter NotebookMIT LicenseMIT

awesome-causality-data

An index of datasets that can be used for learning causality.

Please cite our survey if this data index helps your research.

@article{guo2018survey,
title={A Survey of Learning Causality with Data: Problems and Methods},
author={Guo, Ruocheng and Cheng, Lu and Li, Jundong and Hahn, P. Richard and Liu, Huan},
journal={arXiv preprint arXiv:1809.09337},
year={2018}
}

Updates coming soon

Datasets for Learning Causal Effects (Causal Inference)

With Back-door Criterion

Standard Datasets for Causal Inference

Standard datasets for learning causal effects comes with each instance in the format of (x,d,y).

IHDP1

How is IHDP1 (setting A) simulated

IHDP2

Twins

Job Training (Lalonde 1986)

ACIC Benchmark

Non-i.i.d. Datasets for Causal Inference

Amazon

Without Back-door Criterion

Datasets with instrumental Variables (IV)

Standard datasets for learning causal effects, each instance has the format of (i,x,d,y).

1980 Census Extract

CPS Extract

Datasets for Regression Discontinuity Design

Population Threshold RDD Datasets

Datasets for Learning Causal Relationships (Causal Discovery)

Distinguishing Cause from Effect

Database with cause-effect pairs

Datasets for Connections to Machine Learning

Datasets with randomized test set for recommendation systems

Name Paper URL
Coat Schnabel, Tobias, Adith Swaminathan, Ashudeep Singh, Navin Chandak, and Thorsten Joachims. "Recommendations as treatments: Debiasing learning and evaluation." arXiv preprint arXiv:1602.05352 (2016). download
Yahoo! R3 download