Week 7

Causal Inference for Urban Data

This week, we will discuss using causal inference to understand urban phenomena. In the last weeks, we used statistical techniques to describe and predict urban data. It is time to get to the “why” by analyzing the between causes and effects. This will put us closer to be able to evaluate and design better interventions and policies in the city.

In our practical exercise, we will build on those ideas to analyze the impact of groceries on restaurant visits in the metropolitan area of Boston using mobility data from Advan.

Prepare

📖 Read some papers and webpages/books about Causal inference in general and for spatial data in particular. Here are some suggestions:

• Causal Inference: The Mixtape

• Causal Inference for the Brave and True

• Applied Causal Inference

• Using Synthetic Controls by Abadie [1]

• Causal Inference for Spatial Treatments by Pollman [2]

• Difference-in-Differences Estimation with Spatial Spillovers by Butts [3]

Participate

🖥️ Lecture 7 - Introduction to Causal Inference in CUS

Perform

⌨️ Lab 7 - Causal Inference for Spatial Data

Back to course schedule ⏎

References

[1]

A. Abadie, “Using Synthetic Controls: Feasibility, Data Requirements, and Methodological Aspects,” Journal of Economic Literature, vol. 59, no. 2, pp. 391–425, Jun. 2021, doi: 10.1257/jel.20191450.

[2]

M. Pollmann, “Causal Inference for Spatial Treatments.” arXiv, Jan. 2023. doi: 10.48550/arXiv.2011.00373.

[3]

K. Butts, “Difference-in-Differences Estimation with Spatial Spillovers.” arXiv, Jun. 2023. doi: 10.48550/arXiv.2105.03737.