ddsimca: Data Driven SIMCA

In mdatools: Multivariate Data Analysis for Chemometrics

Data Driven SIMCA

Description

ddsimca is used to develop DD-SIMCA (Data Driven SIMCA) model for one-class classification.

Usage

ddsimca(
  x,
  classname,
  ncomp = min(nrow(x) - 1, ncol(x) - 1, 20),
  center = TRUE,
  scale = FALSE,
  pcv = list("ven", 10),
  alpha = 0.05,
  gamma = 0.01,
  exclrows = NULL,
  exclcols = NULL,
  prep = NULL,
  do.round = TRUE,
  ...
)

Arguments

x	a numerical matrix with data values.
classname	short text (up to 20 symbols) with class name.
ncomp	maximum number of components to calculate.
center	logical, do mean centering of data or not.
scale	logical, do standardization of data or not.
pcv	Procrustes cross-validation settings (see details).
alpha	significance level for making the predictions (can be also adjusted when model is applied to data).
gamma	significance level for detection of outliers (can be also adjusted when model is applied to data).
exclrows	rows to be excluded from calculations (numbers, names or vector with logical values)
exclcols	columns to be excluded from calculations (numbers, names or vector with logical values)
prep	optional list with preprocessing methods created using 'prep' function.
do.round	logical, round or not DoF for distances.
...	any other parameters suitable for pca method.

Details

DD-SIMCA is based on PCA model with additional functionality, so ddsimca class inherits most of the functionality of pca class.

In order to make a decision, DDSIMCA uses score and orthogonal distances to PCA model. It combines the two distances to joint full distance and uses chi-distribution for finding a critical value which is employed as decision rule. More details about DD-SIMCA can be found in [1] (open access).

Procrustes cross-validation (PCV) is used to generate a validation set in order to find optimal model complexity (number of components). The PCV settings are similar to the ones used for conventional cross-validation. The best way is to set 'pcv' value to a list, for example: pcv = list('ven', nseg) for systematic splits or pcv = list('rand', nseg) for random splits. In case if full cross-validation must be employed, use pcv = list('loo').

Value

Returns an object of ddsimca class with following fields:

classname	a short text with class name.
calres	an object of class simcares with classification results for a calibration data.
pvres	an object of class simcares with classification results for a Procrustes validation set.

Fields, inherited from pca class:

ncomp	number of components included to the model.
ncomp.selected	selected (optimal) number of components.
loadings	matrix with loading values (nvar x ncomp).
eigenvals	vector with eigenvalues for all existent components.
expvar	vector with explained variance for each component (in percent).
cumexpvar	vector with cumulative explained variance for each component (in percent).
info	information about the model, provided by user when build the model.

Author(s)

Sergey Kucheryavskiy ([e-mail])

References

1. Kucheryavskiy S, Rodionova O, Pomerantsev A. A comprehensive tutorial on Data-Driven SIMCA: Theory and implementation in web. Journal of Chemometrics. 2024; 38(7):e3556. doi:10.1002/cem.3556

2. S. Kucheryavskiy, O. Rodionova, A. Pomerantsev, Procrustes cross-validation of multivariate regression models. Analytica Chimica Acta. 2023; 1255:341096. doi:10.1016/j.aca.2023.341096.

See Also

Methods for ddsimca objects:

print.ddsimca	shows information about the object.
summary.ddsimca	shows summary statistics for the model.
plot.ddsimca	makes an overview of DD-SIMCA model with four plots.
predict.ddsimca	applies DD-SIMCA model to a new data.

Methods, inherited from classmodel class:

plotPredictions.classmodel	shows plot with predicted values.
plotSensitivity.classmodel	shows sensitivity plot.
plotSpecificity.classmodel	shows specificity plot.
plotMisclassified.classmodel	shows misclassified ratio plot.

Methods, inherited from pca class:

selectCompNum.pca	set number of optimal components in the model
plotScores.pca	shows scores plot.
plotLoadings.pca	shows loadings plot.
plotVariance.pca	shows explained variance plot.
plotCumVariance.pca	shows cumulative explained variance plot.

Examples

## make a SIMCA model for Iris setosa class with full cross-validation
library(mdatools)

data = iris[, 1:4]
class = iris[, 5]

# take first 20 objects of setosa as calibration set
se = data[1:20, ]

# make SIMCA model and apply to test set
model = ddsimca(se, "setosa", pcv = list("ven", 10))
model = selectCompNum(model, 1)

# show information, summary and plot overview
print(model)
summary(model)
plot(model)

mdatools documentation built on March 6, 2026, 5:08 p.m.