The statistical analysis of geochemical compositions

  • John Aitchison


The analysis and interpretation of compositional data, such as major oxide compositions of rocks, has been traditionally plagued by the so-called constant-sum or closure problem. Particular difficulties have been the lack of a satisfactory, interpretable covariance structure and of rich, tractable, parametric classes of distributions on the simplex sample space. Consideration of logistic and logratio transformations between the simplex and Euclidan space has allowed the introduction of new concepts of covariance structure and of classes of logistic-normal distributions which have now opened up a substantial and meaningful array of statistical methodology for compositional data. From the motivation of a wide variety of practical geological problems we examine the range of possibilities with this new approach to the constant-sum problem.

Key words

Closure closed number system logistic logratio geochemistry 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Aitchison, J., 1981a, A new approach to null correlations of proportions: Jour. Math. Geol., v. 13, p. 175–189.Google Scholar
  2. Aitchison, J., 1981b, Distributions on the simplex for the analysis of neutrality,in, C. Taillie, G. P. Patil, and B. Baldessari (Eds.), Statistical distributions in scientific work: D. Reidel, Dordrecht, Holland, p. 147–156.Google Scholar
  3. Aitchison, J., 1982, The statistical analysis of compositional data (with discussion): Jour. Roy. Stat. Soc. Ser. B., v. 44, p. 139–177.Google Scholar
  4. Aitchison, J., 1983, Principal component analysis of compositional data: Biometrika, v. 70, p. 57–65.Google Scholar
  5. Aitchison, J., 1984a, Reducing the dimensionality of compositional data sets: Jour. Math. Geol., v. 16, to appear.Google Scholar
  6. Aitchison, J., 1984b, A general class of distributions on the simplex: Jour. Roy. Stat. Soc. Ser. B., v. 46, to appear.Google Scholar
  7. Aitchison, J. and Bacon-Shone, J. H., 1984, A logcontrast approach to experiments with mixtures: Biometrika, v. 71, to appear.Google Scholar
  8. Aitchison, J. and Dunsmore, I. R., 1975, Statistical prediction analysis: Cambridge University Press.Google Scholar
  9. Aitchison, J. and Lauder, I. J., 1984, Kernel density estimation for compositional data: submitted to Appl. Stat.Google Scholar
  10. Aitchison, J. and Li, C. K. T., 1985, A new approach to classification from compositional data: submitted to Jour. Math. Geol.Google Scholar
  11. Aitchison, J. and Shen, S. M., 1980, Logistic-normal distributions: Some properties and uses: Biometrika, v. 67, p. 261–272.Google Scholar
  12. Aitchison, J. and Shen, S. M., 1984, Measurement error in compositional data: Jour. Math. Geol., v. 16, p. 637–650.Google Scholar
  13. Anderson, T. W., 1958, An Introduction to Multivariate Statistical Analysis: John Wiley & Sons, New York.Google Scholar
  14. Barker, D. S., 1978, Magmatic trends on alkali-iron-magnesium diagrams: Amer. Min., v. 63, p. 531–534.Google Scholar
  15. Becker, N. G., 1968, Models for the response of a mixture: Jour. Roy. Stat. Soc. Ser. B., v. 30, p. 349–358.Google Scholar
  16. Becker, N. G., 1978, Models and designs for experiments with mixtures: Aust. Jour. Stat., v. 20, p. 195–208.Google Scholar
  17. Butler, J. C., 1979, Trends in ternary petrologic variation diagrams—fact or fantasy?: Amer. Min., v. 64, p. 1115–1121.Google Scholar
  18. Chayes, F., 1960, On correlation between variables of constant sum: Jour. Geophys. Res., v. 65, p. 4185–4193.Google Scholar
  19. Chayes, F., 1962, Numerical correlation and petrographic variation: Jour. Geol., v. 70, p. 440–452.Google Scholar
  20. Chayes, F., 1971, Ratio correlation: University of Chicago Press, Illinois.Google Scholar
  21. Chayes, F., 1983, Detecting nonrandom associations between proportions by tests of remaining-space variables: Jour. Math. Geol., v. 15, p. 197–206.Google Scholar
  22. Chayes, F. and Kruskal, W., 1966, An approximate statistical test for correlations between proportions: Jour. Geol., v. 74, p. 692–702.Google Scholar
  23. Connor, R. J. and Mosimann, J. E., 1969, Concepts of independence for proportions with a generalization of the Dirichlet distribution: Jour. Amer. Stat. Assoc., v. 64, p. 194–206.Google Scholar
  24. Cornell, J. A., 1981, Experiments with mixtures: John Wiley & Sons, New York.Google Scholar
  25. Cox, D. R., 1971, A note on polynomial response functions for mixtures: Biometrika, v. 58, p. 155–159.Google Scholar
  26. Dawid, A. P., 1976, Properties of diagnostic data distributions: Biometrics, v. 32, p. 647–658.Google Scholar
  27. Iddings, J. P., 1903, Chemical composition of igneous rocks: U.S. Geol. Survey Prof. Pap. 18.Google Scholar
  28. James, I. R., 1981, Distributions associated with neutrality properties for random proportions,in, C. Taillie, G. P. Patil, and B. Baldessari (Eds.), Statistical distributions in scientific work: D. Reidel, Dordrecht, Holland, p. 125–136.Google Scholar
  29. Kork, J. O., 1977, Examination of the Chayes-Kruskal procedure for testing correlations between proportions: Jour. Math. Geol., v. 9, p. 543–562.Google Scholar
  30. Krumbein, W. C., 1962, Open and closed number systems in stratigraphic mapping: Bull. Amer. Assoc. Pet. Geol., v. 46, p. 2229–2245.Google Scholar
  31. Lauder, I. J., 1978, Computational problems in predictive diagnosis: Compstat 1978, p. 186–192.Google Scholar
  32. Le Maitre, R. W., 1968, Chemical variation within and between volcanic rock series—a statistical approach: Jour. Pet., v. 9, p. 220–252.Google Scholar
  33. Le Maitre, R. W., 1982, Numerical petrography: Elsevier, Amsterdam.Google Scholar
  34. McAlister, D., 1879, The law of the geometric mean: Proc. Roy. Soc., v. 29, p. 367.Google Scholar
  35. Miesch, A. T., 1969, The constant sum problem in geochemistry,in D. F. Merriam (Ed.), Computer Applications in the earth sciences: Plenum Press, New York, p. 161–167.Google Scholar
  36. Morrison, D. F., 1976, Multivariate statistical methods: New York, McGraw-Hill.Google Scholar
  37. Pearson, K., 1897, Mathematical contributions to the theory of evolution. On a form of spurious correlations which may arise when indices are used in the measurement of organs: Proc. Roy. Soc., v. 60, p. 489–498.Google Scholar
  38. Reyment, R., 1977, Presidential address to the International Association for Mathematical Geology: Jour. Math. Geol., v. 9, p. 451–454.Google Scholar
  39. Sarmanov, O. V. and Vistelius, A. B., 1959, On the correlation of percentage values: Dokl. Akad. Nauk. SSSR, v. 126, p. 22–25.Google Scholar
  40. Stephens, M. A., 1974, EDF statistics for goodness of fit and some comparisons: Jour. Amer. Stat. Assoc., v. 69, p. 730–737.Google Scholar
  41. Webb, W. M. and Briggs, L. I., 1966, The use of principal component analysis to screen mineralogical data: Jour. Geol., v. 74, p. 716–720.Google Scholar
  42. Wilks, S. S., 1938, The large-sample distribution of the likelihood ratio for testing composite hypotheses: Ann. Math. Stat., v. 9, p. 60–62.Google Scholar
  43. Zukav, G., 1979, The dancing Wu-Li masters: Bantam, New York.Google Scholar

Copyright information

© Plenum Publishing Corporation 1984

Authors and Affiliations

  • John Aitchison
    • 1
  1. 1.Department of StatisticsUniversity of Hong KongHong Kong

Personalised recommendations