ABSTRACT
For non-stationary processes, a common method to estimate the spatially varying intensity function involves kernel density estimation. Usually, kernel estimation methods focus on estimating the probability density function f(·) rather than the intensity function λ(·). The density function defines the probability of observing an event at a location x and integrates to one across the area of study. In contrast, the intensity function provides the number of events expected per unit area at location x and integrates to the overall mean number of events per unit area. The density() function of spatstat can be used to obtain a kernel estimate of the intensity of a point pattern. The density() function returns an intensity estimate object of class im that can be plotted. The kernel bandwidth used in density() can be extracted with the sigma attribute of the returned intensity object.
