Parametric Procedures in the Analysis of Replicated Pairwise Interaction Point Patterns

A parametric approach fits particular classes of parametric models to the data, uses the model parameter estimates as summaries and tests for differences between groups by comparing fits with and without the assumption of common parameter values across groups. The paper discusses how a parametric approach can be implemented in the specific context of a single-factor replicated spatial experiment and uses simulations to show when the parametric approach can be efficient or potentially misleading. An analysis of the spatial distribution of pyramidal neurons in human patients is also shown.