Back to overview Show all

Other publication (non peer-review)

Book Flexible estimation of spatio-temporal interaction in a point process model for infectious disease spread
Publisher Statistical Modelling Society, Göttingen Germany


Case reports from infectious disease surveillance with registered location and time of infection allow for spatio-temporal point process models of infectious disease spread. An endemic component describes the baseline risk of infection driven by population density as well as temporal and exogenous effects. A second, epidemic component captures interaction between cases and includes covariate effects on the force of infection. Here we investigate nonparametric estimation of spatial as well as temporal interaction using B-splines. Such flexible formulations disclose the distance-decay and time-course of infectivity in a more data-driven manner than previously used parametric models.