Learning about Spatial and Temporal Proximity using Tree-Based Methods

TL;DR

This paper demonstrates that tree-based methods effectively analyze complex spatial and temporal proximity relationships with outcomes, offering a flexible, data-driven alternative to traditional regression techniques.

Contribution

It introduces the application of tree-based methods to study proximity effects on social and political outcomes, highlighting their advantages over conventional regression approaches.

Findings

Tree-based methods reveal significant associations between proximity to border crossings and immigration reform support.

Proximity to mass shootings correlates with support for gun control.

Tree methods outperform traditional regression in modeling proximity-outcome relationships.

Abstract

Learning about the relationship between distance to landmarks and events and phenomena of interest is a multi-faceted problem, as it may require taking into account multiple dimensions, including: spatial position of landmarks, timing of events taking place over time, and attributes of occurrences and locations. Here I show that tree-based methods are well suited for the study of these questions as they allow exploring the relationship between proximity metrics and outcomes of interest in a non-parametric and data-driven manner. I illustrate the usefulness of tree-based methods vis-\`a-vis conventional regression methods by examining the association between: (i) distance to border crossings along the US-Mexico border and support for immigration reform, and (ii) distance to mass shootings and support for gun control.