Tests of Discordancy for Samples from Fisher's Distribution on the Sphere

Abstract

Summary Several tests of discordancy for an outlying value in a sample of data from Fisher's distribution are investigated, and recommendations made about their use. One of the tests is modified for application to testing discordancy of several outliers in a single Fisher sample. The various tests are applied to some samples of artificial data and palaeomagnetic data.

New paleomagnetic data from the northern Puna and western Cordillera Oriental, Argentina: a new insight on the timing of rotational deformation

“…In addition, the site-mean directions corresponding to the Cerro León Grande dacitic dome and to the Chinchillas ignimbrite were not considered in the following analysis because they are markedly removed from the remaining site means. It was determined through the application of a statistical test (Fisher et al, 1981) that such site-mean directions do not belong to the population of site-mean directions obtained for the Laguna de Pozuelos basin (Table 3).…”

Is the Fisher Distribution Additive?

“…From the practical viewpoint it is important whether the difference can be found by statistical means, i.e., by goodness-of-fit tests. The following experiment was carried out in this respect: 1000 samples from distributions ( ) { } 0, 0 ,5 F and , respectively, independent of each other, were simulated after Fisher et al (1981) (Figs. 3 and 4) and combined according to product (4) (Fig.…”

Palaeomagnetic correlation of the Acraman impact structure and the Late Proterozoic Bunyeroo ejecta horizon, South Australia

“…Several such tests in the literature are discussed by Fisher et al (1981) and a more recent test is given by McFadden (1982). The test favoured by Fisher et al (1981) for distributions with precision parameters >10 is denoted E n . This yields a value of 1.148 which is much less than the 0.05 cut-off for F2,i6 equal to 3.63.…”