This is a class representation of a PCA result
Creating Objects
new("pcaRes", scores=[the scores], loadings=[the loadings],
nPcs=[amount of PCs], R2cum=[cumulative R2], nObs=[amount of
observations], nVar=[amount of variables], R2=[R2 for each
individual PC], sDev=[stdev for each individual PC],
centered=[was data centered], center=[original means],
varLimit=[what variance limit was exceeded], method=[method used to
calculate PCA], missing=[amount of NAs],
completeObs=[estimated complete observations])
Slots
"matrix", the calculated scores
"matrix", the calculated loadings
"numeric", the cumulative R2 values
"numeric", the individual standard deviations of the score vectors
"numeric", the individual R2 values
"numeric", cross-validation statistics
"numeric", number of observations
"numeric", number of variables
"logical", data was centered or not
"numeric", the original variable centers
"logical", data was scaled or not
"numeric", the original variable scales
"numeric", the exceeded variance limit
"numeric", the number of calculated PCs
"character", the method used to perform PCA
"numeric", the total amount of missing values in original data
"matrix", the estimated complete observations
"nlpcaNet", the network used by non-linear PCA
Methods (not necessarily exhaustive)
Print function
Extract information about PC relevance
Plot a barplot of standard deviations for PCs
Make a side by side score and loadings plot
Get the number of PCs
Get the number of observations
Cross-validation statistics
Get the number of variables
Get the loadings
Get the scores
Get the dimensions (number of observations, number of features)
Get a logical indicating if centering was done as part of the model
Get the averages of the original variables.
Get the imputed data set
Get a string naming the used PCA method
Get the standard deviations of the PCs
Get a logical indicating if scaling was done as part of the model
Get the scales of the original variablesb
Get the cumulative R2
Henning Redestig
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.