corDiss: Correlation and moving correlation dissimilarity measurements (corDiss)

Description

Computes correlation and moving correlation dissimilarity matrices.

Usage

corDiss(Xr, X2 = NULL, ws = NULL, 
        center = TRUE, scaled = TRUE)

Arguments

a matrix (or data.frame) containing the (reference) data.

an optional matrix (or data.frame) containing data of a second set of observations (samples).

for moving correlation dissimilarity, an odd integer value which specifies the window size. If ws = NULL, then the window size will be equal to the number of variables (columns), i.e. instead moving correlation, the normal correlation will be

center

a logical indicating if the spectral data Xr (and X2 if specified) must be centered. If X2 is specified the data is scaled on the basis of $Xr \cup X2$.

scaled

a logical indicating if Xr (and X2 if specified) must be scaled. If X2 is specified the data is scaled on the basis of $Xr \cup X2$.

Value

a matrix of the computed dissimilarities.

Details

The correlation dissimilarity $cd$ between two obsvervations $x_i$ and $x_j$ is computed as follows:

c d (x_{i}, x_{j}) = \frac{1}{2} (1 - c o r (x_{i}, x_{j}))

The avobe formlula is used when ws = NULL. On the other hand (when ws != NULL) the moving correlation dissimilarity $mcd$ between two obsvervations $x_i$ and $x_j$ is computed as follows:

m c d (x_{i}, x_{j}) = \frac{1}{2 w s} \sum_{k = 1}^{p - w s} (1 - c o r (x_{i, (k : k + w s)}, x_{j, (k : k + w s)}))

where $ws$ represents a given window size which rolls sequantially fom 1 up to $p - ws$ and $p$ is the number of variables of the observations. The function does not accept input data containing missing values.

Examples

Run this code

require(prospectr)
data(NIRsoil)

Xu <- NIRsoil$spc[!as.logical(NIRsoil$train),]
Xr <- NIRsoil$spc[as.logical(NIRsoil$train),]

corDiss(Xr = Xr)

corDiss(Xr = Xr, X2 = Xu)

corDiss(Xr = Xr, ws = 41)

corDiss(Xr = Xr, X2 = Xu, ws = 41)