Testing the complete spatial randomness by Diggle's test without an arbitrary upper limit

摘要

Diggle's test for complete spatial randomness of a given point pattern uses the discrepancy between the estimated and the theoretical form of a summary function as the test statistic. One commonly used discrepancy measure is the supremum of the pointwise differences over a suitably chosen range; the upper bound of the range is an arbitrary but sometimes crucial parameter. This paper shows that when we use Ripley's K-function as the summary function, it is possible to avoid using an arbitrary upper bound by using adapted distance dependent intensity estimators.