Testing Unit Vectors for Correlation,
Abstract
Statistical tests with unit vectors have been very useful in paleomagnetism. Beran and Watson (1967) gave a test for serial correlation of unit vectors. In this paper, formulae are given to make this test computationally easy to apply. A test statistic based on the sum of the cosines of the angles between each pair of vectors is provided to test pairwise correlation of unit vectors. Computable formulae are given for the mean and variance in the randomization distribution of the test statistic. A Monte Carlo experiment is quoted to support a normal approximation for its distribution. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1970
- Accession Number
- AD0729148
Entities
People
- G. S. Watson
- J. W. Tukey
- R. J. Epp
Organizations
- Princeton University