osc {mt}

Orthogonal Signal Correction (OSC)

Description

Data pre-processing by orthogonal signal correction (OSC).

Usage

osc(x, ...)

## Default S3 method:
osc(x, y, method="wold",center=TRUE,osc.ncomp=4,pls.ncomp=10,
   tol=1e-3, iter=20,...)

## S3 method for class 'formula'
osc(formula, data = NULL, ..., subset, na.action = na.omit)

Arguments

formula

A formula of the form groups ~ x1 + x2 + ... That is, the response is the grouping factor and the right hand side specifies the (non-factor) discriminators.

data

Data frame from which variables specified in formula are preferentially to be taken.

x

A matrix or data frame containing the explanatory variables if no formula is given as the principal argument.

y

A factor specifying the class for each observation if no formula principal argument is given.

method

A method for calculating OSC weights, loadings and scores. The following methods are supported:

• wold: Original Wold et al approach.

• sjoblom: Sjoblom et al approach.

• wise: Wise and Gallagher approach.

center

A logical value indicating whether the data set should be centred by column-wise.

osc.ncomp

The number of components to be used in the OSC calculation.

pls.ncomp

The number of components to be used in the PLS calculation.

tol

A scalar value of tolerance for OSC computation.

iter

The number of iteration used in OSC calculation.

...

Arguments passed to or from other methods.

subset

An index vector specifying the cases to be used in the training sample.

na.action

A function to specify the action to be taken if NAs are found. The default action is na.omit, which leads to rejection of cases with missing values on any required variable. An alternative is na.fail, which causes an error if NA cases are found.

Value

An object of class osc containing the following components:

x

A matrix of OSC corrected data set.

R2

R2 statistics. It is calculated as the fraction of variation in X after OSC correction for the calibration (training) data.

angle

An angle used for checking if scores t is orthogonal to y. An angle close to 90 degree means that orthogonality is achieved in the correction process.

w

A matrix of OSC weights.

p

A matrix of OSC loadings.

t

A matrix of OSC scores.

call

The (matched) function call.

center

A logical value indicating whether the data set has been centred by column-wise.

osc.ncomp

The number of component used in OSC computation.

pls.ncomp

The number of component used in PLS computation.

method

The OSC algorithm used.

Note

This function may be called giving either a formula and optional data frame, or a matrix and grouping factor as the first two arguments.

Author(s)

Wanchang Lin

References

Wold, S., Antti, H., Lindgren, F., Ohman, J.(1998). Orthogonal signal correction of near infrared spectra. Chemometrics Intell. Lab. Syst., 44: 175-185.

Westerhuis, J. A., de Jong, S., Smilde, A, K. (2001). Direct orthogonal signal correction. Chemometrics Intell. Lab. Syst., 56: 13-25.

Sjoblom. J., Svensson, O., Josefson, M., Kullberg, H., Wold, S. (1998). An evaluation of orthogonal signal correction applied to calibration transfer of near infrared spectra. Chemometrics Intell. Lab. Syst.,44: 229-244.

Svensson, O., Kourti, T. and MacGregor, J.F. (2002). An investigation of orthogonal correction algorithms and their characteristics. Journal of Chemometrics, 16:176-188.

Wise, B. M. and Gallagher, N.B. http://www.eigenvector.com/MATLAB/OSC.html.

See Also

predict.osc, osc_wold, osc_sjoblom, osc_wise

Examples

data(abr1)
cl   <- factor(abr1$fact$class)
dat  <- abr1$pos

## divide data as training and test data
idx <- sample(1:nrow(dat), round((2/3)*nrow(dat)), replace=FALSE) 

## construct train and test data 
train.dat  <- dat[idx,]
train.t    <- cl[idx]
test.dat   <- dat[-idx,]        
test.t     <- cl[-idx] 

## build OSC model based on the training data
res <- osc(train.dat, train.t, method="wise", osc.ncomp=2, pls.ncomp=4)
names(res)
res
summary(res)

## pre-process test data by OSC
test.dat.1 <- predict(res,test.dat)$x