PLS-discriminant analysis with cross-validation
function [model] = plsda(x,y_class,split,lv,(metric),(scale),(threshold))
a matrix (n x q) or a Div structure
classification of the observations; a disjunctive matrix (n x nclass) or a conjunctive vector (n x 1) or Div structures
for the cross-validation; a number of blocks, or a vector of dimensions (n x 1) identifying by a number the CV group of each observation
the number of latent variables or eigenvectors used to build the models
the metric used to measure the distance of an observation scores to a class
metric=0: Mahalanobis distance (by default)
metric=1: usual Euclidian distance
the columns of x are always centered; scale defines the standardization
scale='c': no standardization
scale='cs': standardisation; the columns of x are divided by their standard deviation (by default)
the lowest value for attributing an observation to a class (by default= 1/nclass )
the confusion matrix obtained with the calibration models,expressed as numbers of observations
model.conf_cal.d is an hyper-matrix of dimensions (nclass x nclass x lv)
the confusion matrix obtained with the calibration models, expressed as percentages
model.conf_cal.d is an hyper-matrix of dimensions (nclass x nclass x lv)
the confusion matrix obtained with the cross-validations, expressed as percentages
model.conf_cv.d is an hypermatrix of dimensions (nclass x nclass x lv)
the percentage of classification errors for calibrations (1st column) and cross-validations (2nd column)
model.err.d is a matrix (lv x 2)
the percentage of calibration error, for each class
model.errbycl_cal.d is a matrix (lv x nclass)
the percentage of cross-validation error, for each class
model.errbycl_cv.d is a matrix (lv x nclass)
the percentage of not-classed observations (all predictions lower than the threshold)
model.notclassed.d is a vector (lv x 1)
the percentage of not-classed observations (all predictions lower than the threshold), for each class
model.notclassed_bycl.d is a matrix (lv x nclass)
the discriminant method; here: 'plsda'
the eigenvectors
model.loadings.d is a matrix of dimensions (q x lv)
the metrix associated to the loadings; for 'plsda': inv(x'x)
model.model_metric.d is a matrix of dimensions (q x q)
the metric used onto the scores for the calculation of distances between observations and groups
the applied standardisation
the applied threshold