correlationCoefPlot {GCDkit} | R Documentation |
Produces set of plots of correlation coefficient patterns, for the entire dataset or separately for each group.
correlationCoefPlot(elems = NULL, groups = NULL, method = "pearson", use = "pairwise.complete.obs")
elems |
list of desired elements |
groups |
a character vector with grouping information |
method |
a character string indicating which correlation coefficient (or covariance) is to be computed |
use |
an optional character string giving a method for computing covariances in the presence of missing values |
This function calculates and plots correlation coefficient patterns either for
the whole dataset (if 'groups
= NULL') or by individual groups.
For meaning of parameters 'methods
' and 'use
', see cor
.
The utility of pairwise correlation coefficient patterns was demonstrated by Rollinson (1993 and references therein). In principle, a similarity in correlation patterns between two or more elements means their analogous geochemical behaviour, potentially reflecting the operation of the same petrogenetic/geological process (fractional crystallization, partial melting, weathering, hydrothermal alteration...)
The variables are selected using the function 'selectColumnsLabels
'.
A list with the values of pairwise correlation in the whole dataset or in each of the groups.
Vojtěch Janoušek, vojtech.janousek@geology.cz
Rollinson HR (1993) Using Geochemical Data: Evaluation, Presentation, Interpretation. Longman, London, p. 1-352
data(atacazo) accessVar("atacazo") # The whole dataset correlationCoefPlot(elems="K,Rb,Ba,Sr,Cr,Ni,Zr,Nb,Ti") # By currently defined groups groupsByLabel("Volcano") correlationCoefPlot(elems="K,Rb,Ba,Sr,Cr,Ni,Zr,Nb,Ti",groups)