The importance of managing co-variate and distance biases in point-pattern clustering. GeoComputation 2019
conference contributionposted on 2019-12-01, 23:14 authored by Peter Whigham, Brandon de Graaf, Rashi Srivastava, Paul Glue
Assessing social conditions for individuals often involves the evaluation of point-based patterns over some geographic region. This paper demonstrates the influence of co-variate information on observed clustering patterns and a method for assessing the role of distance metrics on clustering. The approaches are applied to assess social contagion of deliberate self-harm for a set of individuals observed over a two year period in a small community in New Zealand.