smacof (version 1.8-13)

uniscale: Unidimensional Scaling

Description

Simple implementation where all dissimilarity permutations are subject to a 1D MDS fit and the one which leads to a minimal stress values is returned.

Usage

uniscale(delta, weightmat = NULL)

Arguments

delta

Either a symmetric dissimilarity matrix or an object of class "dist"

weightmat

Optional matrix with dissimilarity weights

Value

delta

Observed dissimilarities, not normalized

confdiss

Configuration distances

conf

Vector with fitted configurations

stress

Stress-1 value

weightmat

Weight matrix

nobj

Number of objects

npermtot

Total number of permutations (factorial)

npermscale

Number of accepted permutations (monotonicity check)

References

Mair P., De Leeuw J. (2015). Unidimensional scaling. In Wiley StatsRef: Statistics Reference Online, Wiley, New York.

See Also

mds

Examples

## unidimensional scaling of Plato's 7 works
PlatoD <- dist(t(Plato7))
fit.uni <- uniscale(PlatoD)
fit.uni
plot(fit.uni)