Open AccessNew statistical methods for directional data—I. Bootstrap comparison of mean directions and the fold test in palaeomagnetism
Author(s) -
Fisher Nicholas I.,
Hall Peter
Publication year - 1990
Publication title -
geophysical journal international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.302
H-Index - 168
eISSN - 1365-246X
pISSN - 0956-540X
DOI - 10.1111/j.1365-246x.1990.tb06570.x
Subject(s) - fold (higher order function) , paleomagnetism , statistics , sample size determination , statistical hypothesis testing , sample (material) , mathematics , sample mean and sample covariance , computer science , geology , geophysics , estimator , programming language , chemistry , chromatography
SUMMARY This paper, and a companion paper, present some new computer‐intensive statistical methods for handling small‐sample problems in directional data analysis in areas where existing methods are not valid. Here, we describe ‘bootstrap’ tests for comparing the the mean directions of two or more samples of 2‐D or 3‐D unit vectors. A case of particular interest is the so‐called fold test in palaeomagnetism, a situation in which sample sizes tend to be small, and sample dispersions different. Bootstrap methods give the first satisfactory method for handling this problem. Examples are given for both circular and spherical data.