Tests for isotropy in spatial point patterns – A comparison of statistical indices

Isotropy of a point process, defined as invariance of the distribution\ua0under rotation, is often assumed in spatial statistics. Formal\ua0tests for the hypothesis of isotropy can be created by comparing\ua0directional summary statistics in different directions. In this paper,\ua0the statistical powers of tests based on a variety of summary\ua0statistics and several choices of deviance measures are compared\ua0in a simulation study. Four models for anisotropic point processes\ua0are considered covering both regular and clustered cases.\ua0We discuss the robustness of the results to changes of the tuning\ua0parameters, and highlight the strengths and limitations of the\ua0methods