The importance of managing co-variate and distance biases in point-pattern clustering. GeoComputation 2019
conference contributionposted on 01.12.2019 by Peter Whigham, Brandon de Graaf, Rashi Srivastava, Paul Glue
Any type of content contributed to an academic conference, such as papers, presentations, lectures or proceedings.
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.