View source: R/pco_selection.R
PCOcutoff | R Documentation |
Compares eigenvalue distribution of PCO to that with randomized data in order to extract axes with significant signal.
PCOcutoff(data, nreps, metric)
data |
Raw data. |
nreps |
Number of iterations. |
metric |
Distance matrix calculation metric. "euclidean", "manhattan", or "gower". |
Significant PCO's are defined as those with greater eigenvalues than from random data. Therefore the PCO cutoff is axes whose eigenvalues fall below the mean eigenvalue for that axis from the randomized data. Data are randomly sampled by row.
Warning: Bootstrapping is sensitive to unequale variances of columns. You may want to use
scale
to normalize before using this approach
Eigenvalues of original PCO (eigen.true), mean eigenvalues from the randomized data (eigen.mean), standard deviation of randomized eigenvalues (eigen.sd), PCOs with eigenvalues exceeding the randomized data (sigpco), matrix of all eigenvalues produced by randomization (eigen.boot).
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.