createSymDist4PTCA: Create a matrix of the symmetric difference distance /...

In HerveAbdi/PTCA4CATA: Partial Triadic Analysis (PTCA) for Check All That Apply (CATA) Data

Create a matrix of the symmetric difference distance / cross-products for the 3rd dimension of a brick of non-negative numbers (i.e., a brick of data for a BeTA or a CATA test).

Description

createSymDist4PTCA creates a matrix of the symmetric difference distance / cross-products between a set of K matrices stored as the 3rd dimension of an I*J*K brick of non-negative numbers (i.e., a brick of data for a BeTA/CATA test). The symmetric difference distance between two (0/1) matrices X and Y is computed as sum(abs(X - Y)). The K*K distance matrix is converted using double centering into a cross-product matrix that can be analyzed by an eigen-decomposition to provide a STATIS-like scalar-product/RV-map.

Usage

createSymDist4PTCA(dataCube)

Arguments

dataCube

an I*J*K brick of non-negative numbers (i.e., a brick of data for a CATA test)

Details

Note that the symmetric difference distance is meaningful only between 0/1 matrices.

Value

return.list A list with Distance: a K*K (symmetric difference) distance matrix CrossProduct: a K*K cross-product matrix obtained by double centering the distance matrix (the cross-product can be eigen-decomposed to give an MDS).

Author(s)

Herve Abdi

Examples

## Not run: 
if(interactive()){
 # with aCubeOfCATAData being an I*J*K array
 DistanceMat <- createSymDist4PTCA(aCubeOfCATAData)
 }

## End(Not run)

HerveAbdi/PTCA4CATA documentation built on July 17, 2022, 5:41 a.m.