Bayesian method for inferring the impact of geographical distance on intensity of communication

TL;DR

This paper introduces a Bayesian LASSO-based statistical model to analyze how communication intensity declines with geographical distance, allowing for different rates at short and long distances, with applications in social and spatial analysis.

Contribution

The paper presents a novel Bayesian model that captures discontinuities in communication decay rates over distance, improving understanding of spatial communication patterns.

Findings

Communication intensity declines with distance.

Decay rates differ for short and long distances.

Model applied successfully to mobile phone data.

Abstract

Both theoretical models and empirical findings suggest that the intensity of communication among groups of people declines with their degree of geographical separation. There is some evidence that rather than decaying uniformly with distance, the intensity of communication might decline at different rates for shorter and longer distances. Using Bayesian LASSO for model selection, we introduce a statistical model for estimating the rate of communication decline with geographic distance that allows for discontinuities in this rate. We apply our method to an anonymized mobile phone communication dataset. Our results are potentially useful in settings where understanding social and spatial mixing of people is important, such as in cluster randomized trials design.