A two-stage Cox process model with spatial and nonspatial covariates

• NCJ Number: 308128
• Author(s): Claire Kelling; Murali Haran
• Date Published: October 2022
• Grant Number(s): 2020-R2-CX-0033
• Agencies: NIJ-Sponsored
• Publication Type: Report (Study/Research)

Annotation

This study presents a new two-stage Cox process model to consider disparate problems such as police use of force incidents and forest fires.

Abstract

Rich new marked point process data allow researchers to consider disparate problems such as the factors affecting the location and type of police use of force incidents, and the characteristics that impact the location and size of forest fires. The authors develop a two-stage log Gaussian Cox process that models these data in terms of both spatial (community-level) and nonspatial (individual or event-level) characteristics; both types of covariates are present in the examples they consider and are not easy to incorporate via existing methods. Via simulated and real data examples the authors find that their model is easy to interpret and flexible, accommodating multiple types of marks and multiple types of spatial covariates. In the first example authors consider, their approach allows them to study the impact of community-level socioeconomic features such as unemployment as well as event-level features such as officer tenure on force used by police, illustrated through simulated examples. In their second example, the authors consider factors that impact the locations and severity of forest fires from the Castilla-La Mancha region of Spain between 2004–2007. (Published Abstract Provided)