Violent Crime in London: An Investigation using Geographically Weighted Regression

TL;DR

This study investigates the spatial distribution of violent crime in London using Geographically Weighted Regression and clustering, revealing local variations and aiding targeted policy interventions.

Contribution

It combines visual analytics, statistical methods, and GWR to identify demographic factors influencing violent crime with a focus on local variation and effective clustering.

Findings

GWR reveals significant local variation in violent crime rates.

Five distinct clusters of violent crime areas are identified in London.

Human reasoning helps address collinearity among demographic features.

Abstract

Violent crime in London is an area of increasing interest following policing and community budget cuts in recent years. Understanding the locally-varying demographic factors that drive distribution of violent crime rate in London could be a means to more effective policy making for effective action. Using a visual analytics approach combined with Statsitical Methods, demographic features which are traditionally related to Violent Crime Rate (VCR) are identified and OLS Univariate and Multivariate Regression are used as a precursor to GWR. VIF and pearson correlation statistics show strong colinearity in many of the traditionally used features and so human reasoning is used to rectify this. Bandwidth kernel smoothing size of 67 with a Bi-Square type is best for GWR. GWR and OLS regression shows that there is local variation in VCR and K-Means clustering using 5 clusters provides an effective way of seperating violent crime in London into 5 coherent groups.