Partial least squares and compositional data: problems and alternatives

Abstract It is still widely unknown in chemometrics that the statistical analysis of compositional data requires fundamentally different tools than a similar analysis of unconstrained data. This article examines the problems that potentially occur when one performs a partial least squares (PLS) analysis on compositional data and suggests logcontrast partial least squares (LCPLS) as an alternative.

[1]  Felix Chayes,et al.  An Approximate Statistical Test for Correlations between Proportions , 1966, The Journal of Geology.

[2]  F. Chayes.  On correlation between variables of constant sum , 1960 .

[3]  I. Helland.  ON THE STRUCTURE OF PARTIAL LEAST SQUARES REGRESSION , 1988 .

[4]  R. Maitre.  Chemical Variation within and between Volcanic Rock Series—A Statistical Approach , 1968 .

[5]  Cidambi Srinivasan,et al.  Box–Cox transformations in the analysis of compositional data , 1991 .

[6]  Richard A. Reyment.  Multivariate analysis in geoscience: Fads, fallacies and the future , 1987 .

[7]  Nouna Kettaneh-Wold,et al.  Analysis of mixture data with partial least squares , 1992 .

[8]  J. Aitchison.  Principal component analysis of compositional data , 1983 .

[9]  J. A. H. Alkemade,et al.  Multiscale segmentation of well logs , 1992 .

[10]  John C. Butler,et al.  Principal components analysis using the hypothetical closed array , 1976 .

[11]  E. Lukács.  A Characterization of the Gamma Distribution , 1955 .

[12]  A. Höskuldsson.  PLS regression methods , 1988 .

[13]  Null correlation for proportions , 1969 .

[14]  J. Friedman,et al.  A Statistical View of Some Chemometrics Regression Tools , 1993 .

[15]  John Aitchison,et al.  The Statistical Analysis of Compositional Data , 1986 .

[16]  L. E. Wangen,et al.  A theoretical foundation for the PLS algorithm , 1987 .

[17]  R. Reyment.  The statistical analysis of compositional data, by John Aitchison: Chapman & Hall, London, 1986, XV + 416 pages, price £29.95, ISBN 0-412-28060-4 , 1988 .

[18]  A. Woronow,et al.  Chemical changes induced in aragonite using treatments for the destruction of organic material , 1991 .

[19]  R. Maitre.  Petrology of Volcanic Rocks, Gough Island, South Atlantic , 1962 .

[20]  Estimation in compositional data analysis , 1991 .

[21]  W. Windig.  Mixture analysis of spectral data by multivariate methods , 1988 .

[22]  J. Mosimann.  On the compound multinomial distribution, the multivariate β-distribution, and correlations among proportions , 1962 .

[23]  R. Thompson,et al.  Major Element Chemical Variation in the Eocene Lavas of the Isle of Skye, Scotland , 1972 .

[24]  Distinction between Permian and post-Permian igneous rocks in the southern Sydney Basin, New South Wales, on the basis of major-element geochemistry , 1981 .

[25]  M. Degroot,et al.  Bayesian Statistics 2. , 1987 .

[26]  J. Kork.  Examination of the Chayes-Kruskal procedure for testing correlations between proportions , 1977 .

[27]  John Aitchison,et al.  The statistical analysis of geochemical compositions , 1984 .

[28]  Ganapati P. Patil,et al.  Statistical Distributions in Scientific Work , 1981 .

[29]  K. Sjoedin.  Minimizing effects of closure on analytical data , 1984 .