The application of multivariate permutation methods based on distance functions in the earth sciences
- 31 July 1991
- journal article
- Published by Elsevier BV in Earth-Science Reviews
- Vol. 31 (1), 55-71
- https://doi.org/10.1016/0012-8252(91)90042-e
Abstract
A variety of potential applications involving two classes of multivariate permutation methods based on distance functions are presented. These methods avoid the impossible distributional assumptions (knowing the analytic description of real data) associated with classical multivariate parametric methods. The relaxed distributional assumptions of permutation methods yield new statistical analyses which would be exceedingly difficult to formulate in terms of parametric methods. Applications of the first class include detection methods for various types of group difference, single and multiple clumping of rectilinear and cyclic phenomena, regular patterns, and first-order autoregressive occurrences (e.g., greenhouse effects). Applications of the second class include methods for evaluating numerical model output agreement with observed data and multivariate distribution-free linear-model analysis procedures.Keywords
This publication has 31 references indexed in Scilit:
- The verification of numerical models with multivariate randomized block permutation proceduresArchiv für Meteorologie, Geophysik und Bioklimatologie Serie A, 1989
- Geometric consistency for regression model estimation and testing in climatology and meteorologyAtmosphere-Ocean, 1989
- A Generalization of Cohen's Kappa Agreement Measure to Interval Measurement and Multiple RatersEducational and Psychological Measurement, 1988
- F9. L1, L2 and L∞ regression models: Is there a difference?Journal of Statistical Planning and Inference, 1987
- Non-metric statistical analyses: Some metric alternativesJournal of Statistical Planning and Inference, 1986
- Geometric Concerns Pertaining to Applications of Statistical Tests in the Atmospheric SciencesJournal of the Atmospheric Sciences, 1985
- Moment approximations as an alternative to the F test in analysis of varianceBritish Journal of Mathematical and Statistical Psychology, 1983
- THE WEIGHTED MEDIAN AND MULTIPLE REGRESSION1Australian Journal of Statistics, 1983
- Climax I and II: Distortion Resistant Residual AnalysesJournal of Applied Meteorology, 1982
- A class of nonparametric tests based on multiresponse permutation proceduresBiometrika, 1981