Power of Tests for Spatial Randomness in Patterns with Small Number of Events

ABSTRACT A spatial point pattern is a set of point locations distributed by a random process within some region. The sta-tistical analysis of spatial point pattern data usually begins with a test for spatial randomness. The aim of the present work is to investigate the performance of tests for spatial randomness in patterns with small number of events. Tests based on F, G, J and K functions are applied in a large number of non-random patterns generated via computer simulation. The number of times that the test successfully detects non-randomness in non-random patterns gives an indication of its power. It is showed that the typical hierarchy of powers (Po) for patterns of 10 events against aggregation is Po(K)&gt