The two-point angular correlation function is a traditional method used to search for deviations from expectations of isotropy. In this paper we develop and explore a statistically descriptive three-point method with the intended application being th
e search for deviations from isotropy in the highest energy cosmic rays. We compare the sensitivity of a two-point method and a shape-strength method for a variety of Monte-Carlo simulated anisotropic signals. Studies are done with anisotropic source signals diluted by an isotropic background. Type I and II errors for rejecting the hypothesis of isotropic cosmic ray arrival directions are evaluated for four different event sample sizes: 27, 40, 60 and 80 events, consistent with near term data expectations from the Pierre Auger Observatory. In all cases the ability to reject the isotropic hypothesis improves with event size and with the fraction of anisotropic signal. While ~40 event data sets should be sufficient for reliable identification of anisotropy in cases of rather extreme (highly anisotropic) data, much larger data sets are suggested for reliable identification of more subtle anisotropies. The shape-strength method consistently performs better than the two point method and can be easily adapted to an arbitrary experimental exposure on the celestial sphere.
Two separate statistical tests are described and developed in order to test un-binned data sets for adherence to the power-law form. The first test employs the TP-statistic, a function defined to deviate from zero when the sample deviates from the po
wer-law form, regardless of the value of the power index. The second test employs a likelihood ratio test to reject a power-law background in favor of a model signal distribution with a cut-off.
