smacof (version 1.10-8)

inverseMDS: Inverse MDS

Description

Inverse MDS finds the set of all dissimilarity matrices, whose MDS solution is given as input configuration.

Usage

inverseMDS(x)

Arguments

x

Configuration matrix

Value

confdist

Euclidean distance matrix

dissmat

List of dissimilarity matrices

Details

One solution to inverse MDS is simply the Euclidean distances between the points in the low-dimensional space. There will typically be many additional solutions, i.e., many high-dimensional dissimilarity arrangements optimally represented by a given low-dimensional configuration.

References

De Leeuw, J., & Groenen P. J. F. (1997). Inverse multidimensional scaling. Journal of Classification, 12, 3-21.

De Leeuw, J., Groenen P. J. F., & Mair, P. (2016). On Inverse Multidimensional Scaling. http://rpubs.com/deleeuw/148292.

See Also

mds

Examples

# NOT RUN {
## Inverse MDS on theoretical rectangle configuration
ifit <- inverseMDS(rect_constr)   
str(ifit)
# }