Trivariate Kernel Density Estimation of Spatiotemporal Crime Events with Case Study for Lithuania

The paper presents the results of investigation applicability spatiotemporal kernel density estimation (KDE) methods for mapping violent crime in Lithuania. Spatiotemporal research helps to understand and control specific types crime, thereby contributing Sustainable Development Goals. target dataset contained 135,989 records events registered by police Lithuania from 2015–2018 that were classified as violent. focused on choosing appropriate KDE functions their parameters modeling point pattern this particular type crime. aim was estimate density, mass, intensity function(s) so they can be used further confirmatory spatial modeling. application-driven objective obtain reliable practically interpretable surfaces events. Several options improving extending investigated are demonstrated.