Leave-one-out cross-validation for a dudi
Leave-one-out cross-validation to check the dispersion of row coordinates in a dudi.
## S3 method for class 'dudi' loocv(x, progress = FALSE, ...)
x |
the dudi of the |
progress |
logical to display a progress bar during computations (see the |
... |
further arguments passed to or from other methods |
This function returns a list with two elements: $XValCoord and $lsFac, the cross-validated row coordinates and a factor to plot them. The analysis is redone after removing each row of the data table, one at a time. The coordinates of the missing row are computed by projection as supplementary element in the dudi space. This can be used to check the dispersion of the coordinates of one point and it's sensitivity to outliers.
a list with two elements: $XValCoord and $lsFac containing the cross-validated row coordinates and the factor to plot them using the s.class (see example).
Jean Thioulouse
data(meaudret) envpca <- dudi.pca(meaudret$env, scannf = FALSE, nf = 3) xvpca <- loocv(envpca) s.match(envpca$li, xvpca$XValCoord)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.