R/multivariate_analyses.R
exp_var.RdCalculate the percentage of total original variation accounted by axes generated by different multivariate ordination methods.
exp_var(ordination, nax = 10, digits = 5)An object with the results from an ordination method (i.e.,
a "prcomp", "bg_prcomp", "phy_prcomp",
"pls_shapes", "phy_pls_shapes", "burnaby" or
"phy_burnaby" object).
Integer; information about how many axes the user wishes to be returned.
Integer; amount of decimal numbers to be returned.
A table informing the percentages and cumulative percentages of original variation accounted by each ordination axis.
#load data
data("shells")
#extract shapes, log sizes and species classification
species <- shells$data$species
shapes <- shells$shapes$coe
sizes <- log(shells$sizes)
#perform PCA, bgPCA, and PLS
pca <- prcomp(shapes)
bgpca <- bg_prcomp(shapes, groups = species)
pls <- pls_shapes(shapes, X = sizes)
#compare percentages of total variation accounted by the first axes when
#the analysis is unsupervised (PCA), or supervised by species (bgPCA) or
#by log size (PLS)
head(exp_var(pca))
#> variance cummulative
#> PC1 88.65268 88.65268
#> PC2 4.96295 93.61562
#> PC3 2.97799 96.59362
#> PC4 1.24819 97.84181
#> PC5 0.67411 98.51592
#> PC6 0.47215 98.98807
exp_var(bgpca)
#> variance cummulative
#> bgPC1 115.74130 115.7413
#> bgPC2 4.54123 120.2825
#> bgPC3 0.34279 120.6253
#> bgPC4 NA NA
#> bgPC5 NA NA
#> bgPC6 NA NA
#> bgPC7 NA NA
#> bgPC8 NA NA
#> bgPC9 NA NA
#> bgPC10 NA NA
exp_var(pls)
#> variance cummulative
#> PLS-1 7.86911 7.86911
#> PLS-2 NA NA
#> PLS-3 NA NA
#> PLS-4 NA NA
#> PLS-5 NA NA
#> PLS-6 NA NA
#> PLS-7 NA NA
#> PLS-8 NA NA
#> PLS-9 NA NA
#> PLS-10 NA NA