cjerzak/ImageConfounding

---
license: mit
---

**Replication Data for:** Integrating Earth Observation Data into Causal Inference: Challenges and Opportunities

**Details:**
`YandW_mat.csv` contains individual-level observational data. In the dataset, `LONGITUDE` and `LATITUDE` refer to the approximate geo-referenced long/lat of observational units. Experimental outcomes are stored in `Yobs`. The treatment variable is stored in `Wobs`. The unique image key for each observational unit is saved in `UNIQUE_ID`.

Geo-referenced satellite images are saved in
`./Nigeria2000_processed/%s_BAND%s.csv`, where the first "`%s`" refers to the image key associated with each observation (saved in `UNIQUE_ID` in `YandW_mat.csv`) and `BAND%s` refers to one of 3 bands in the satellite imagery.

**Code Link:** https://github.com/cjerzak/causalimages-software/

**Paper Reference:** Connor T. Jerzak, Fredrik Johansson, Adel Daoud. Integrating Earth Observation Data into Causal Inference: Challenges and Opportunities. ArXiv Preprint, 2023. [[PDF]](https://arxiv.org/pdf/2301.12985)

原始信息汇总

数据集概述

数据集名称

Replication Data for: Integrating Earth Observation Data into Causal Inference: Challenges and Opportunities

数据集内容

• YandW_mat.csv: 包含个体级别的观测数据。其中：

  • LONGITUDE 和 LATITUDE: 观测单位的近似地理参考经纬度。
  • Yobs: 实验结果。
  • Wobs: 处理变量。
  • UNIQUE_ID: 每个观测单位的唯一图像键。

• 卫星图像: 存储于 ./Nigeria2000_processed/%s_BAND%s.csv，其中：

  • 第一个 "%s" 对应于 YandW_mat.csv 中的 UNIQUE_ID。
  • BAND%s 表示卫星图像中的一个波段，共有3个波段。

数据集许可证

MIT