Analysis of Distance for Structured Multivariate Data : Review and Illustration

• Autorzy: Wojtek J. Krzanowski

Warianty tytułu

Analiza odległości wielowymiarowych danych strukturalnych : przegląd i przykład

• Języki publikacji: EN

Abstrakty

EN

This has been a necessarily brief review, but the hope is that sufficient detail has been provided for the reader to be able to apply the various analyses that have been outlined to other data sets. However, some cautionary points need to be borne in mind at all stages. Most descriptive analyses rely on just two-dimensional graphical displays, and these should always be first checked for adequacy using any of the standard MDS measures (e.g. STRESS, proportion of eigenvalues to trace, etc). The variability inherent in multidimensional scalings should also be remembered, and when comparing two MDS diagrams the arbitriness of signs of axes should be allowed for (see [Krzanowski 2006] for suggestions). In breaking effects down into single degrees of freedom, care should be taken over choosing appropriate H matrices. Finally, if biplots are computed it should be remembered that the linearity is only an approximation; residual sums of squares from the least squares fitting will give an indication as to how close the approximation really is. If these cautionary points are attended to, then the methods surveyed above should provide useful analyses in many situations. (fragment of text)

Wiele wielowymiarowych zbiorów danych ma strukturę sugerującą wykorzystanie analizy MANOVA, jednakże nie spełniają jej wymaganych założeń. Od niedawna zainteresowanie wzbudza analogiczny typ analizy, przeprowadzanej jednak na macierzy niepodobieństw otrzymanej z nieprzetworzonych danych. Niniejsza praca zawiera przegląd podstaw takiego podejścia, pokazuje główne etapy takiej analizy. Zaprezentowano również, w jaki sposób biploty i analiza wrażliwości mogą być dodane do metody podstawowej, a także ilustruje przedstawione pojęcia za pomocą zbioru danych pochodzących z ekologii. (abstrakt oryginalny)

Słowa kluczowe

EN

Distance measures; Data analysis

PL

Miara odległości; Analiza danych

• Czasopismo: Prace Naukowe Akademii Ekonomicznej we Wrocławiu. Taksonomia
• Rocznik: 2006
• Tom: 13
• Numer: nr 1126 Klasyfikacja i analiza danych - teoria i zastosowania
• Strony: 15--26

Twórcy

autor

Wojtek J. Krzanowski

• University of Exeter Business School, United Kingdom

Bibliografia

• Anderson M.J. (2001), Permutation Tests for Univariate or Multivariate Analysis of Variance and Regression, "Canadian Journal of Fisheries and Aquatic Sciences" 58, p. 626-639.
• Anderson M.J., Willis T.J. (2003), Canonical Analysis of Principal Coordinates: A Useful Method of Constrained Ordination for Ecology, "Ecology" 84, p. 511-525.
• Bray J.R., Curtis J.T. (1957), An Ordination of the Upland Forest Communities of Southern Wisconsin, "Ecological Monographs" 27, p. 325-349.
• Cuadras C.M., Arenas C. (1990), A Distance Based Regression Model for Prediction with Mixed Data, "Communications in Statistics - Theory and Methods" 19, p. 2261-2279.
• d'Aubigny G.D. (1988), The Additive Decomposition of Some Entropy Functions and (Constrained) Ordination Methods, Proceedings of the XIVth International Biometric Conference, p. 455-485.
• Deleeuw J., Meulman J. (1986), A Special Jackknife for Multidimensional Scaling, "Journal of Classification" 3, p. 97-112.
• Digby P.G.N., Gower J.C. (1981), Ordination between- and within-Groups Applied to Soil Classification, [w:] Down-to-Earth Statistics: Solutions Looking for Geological Problems, ed. D.F. Merriam, p. 63-75, Syracuse University Geological Contribution, New York.
• Gower J.C. (1966), Some Distance Properties of Latent Root and Vector Methods Used in Multivariate Analysis, "Biometrika" 53, p. 325-338.
• Gower J.C. (1968), Adding a Point to Vector Diagrams in Multivariate Analysis, "Biometrika" 55, p. 582-585.
• Gower J.C. (1985), Measures of Similarity, Dissimilarity and Distance, [w:] Encyclopedia of Statistical Sciences, vol. 5, edp. S. Kotz, N.L. Johnson, C.B. Read, John Wiley and Sons, New York.
• Gower J.C., Hand D.J. (1996), Biplots, Chapman and Hall, London.
• Gower J.C., Krzanowski W.J. (1999), Analysis of Distance for Structured Multivariate Data and Extensions to Multivariate Analysis of Variance, "Applied Statistics" 48, p. 505-519.
• Krzanowski W.J. (2002), Multifactorial Analysis of Distance in Studies of Ecological Community Structure, "Journal of Agricultural, Biological and Environmental Statistics" 7, p. 222-232.
• Krzanowski W.J. (2004), Biplots for Multifactorial Analysis of Distance, "Biometrics" 60, p. 517-524.
• Krzanowski W.J. (2006), Sensitivity in Metric Scaling and Analysis of Distance, "Biometrics" 62, in presp.
• Krzanowski W.J., Marriott F.H.C. (1994), Multivariate Analysip. Part 1: Distributions, Ordination and Inference, Kendall's Library of Statistics, Edward Arnold, London.
• Legendre P., Anderson M.J. (1999), Distance-Based Redundancy Analysis: Testing Multispecies Responses in Multifactorial Ecological Experiments, "Ecological Monographs" 69, p. 1-24.
• Manly B.F.J. (1997), Randomization and Monte Carlo Methods in Biology, Chapman and Hall, London.
• Mardia K.V., Kent J.T., Bibby J.M. (1979), Multivariate Analysis, Academic Press, London.
• McArdle B.H., Anderson M.J. (2001), Fitting Multivariate Models to Community Data: A Comment On Distance-Based Redundancy Analysis, "Ecology" 82, p. 290-297.
• Pillar V. De P., Orloci L. (1996), On Randomization Testing in Vegetation Science: Multifactor Comparisons of Releve Groups, , Journal of Vegetation Science" 7, p. 585-592.
• Ramsay J.O. (1982), Some Statistical Approaches to Multidimensional Scaling Data (with Discussion), "Journal of the Royal Statistical Society" Series A, 145, p. 285-312.
• Widdicombe S., Austen M.C. (2001), Interactions between Physical Disturbance and Organic Enrichment: An Important Element in Structuring Benthic Communities, "Limnology and Oceanography" 46, p. 1720-1733.