INDSCAL: weighted MDS with individual differences

Carroll & Chang (1970) weighted MDS with a shared stimulus space and per-individual dimension weights.

Usage

morie_spatial_voting_indscal(
  dissimilarities,
  n_dims = 2L,
  max_iter = 300L,
  tol = 1e-06
)

Arguments

dissimilarities

List of (n_stim x n_stim) dissimilarity matrices.

n_dims

Number of dimensions.

max_iter

Maximum ALS iterations.

tol

Convergence tolerance on configuration change.

Value

List with group_config, weights, stress, iterations, n_individuals, n_stimuli.

References

Carroll, J. D. and Chang, J.-J. (1970). "Analysis of Individual Differences in Multidimensional Scaling via an N-way Generalization of Eckart-Young Decomposition." Psychometrika, 35(3).

Examples

set.seed(1)
D1 <- as.matrix(dist(matrix(rnorm(20), 5)))
D2 <- as.matrix(dist(matrix(rnorm(20), 5)))
morie_spatial_voting_indscal(list(D1, D2), n_dims = 2L, max_iter = 30L)
#> $group_config
#>               [,1]           [,2]
#> [1,] 1.594706e+150 -4.748038e-136
#> [2,] 6.841884e+151 -8.272560e-138
#> [3,] 2.963986e+153 -7.216644e-139
#> [4,] 2.099617e+155 -2.202051e-140
#> [5,] 1.794468e+157 -1.332289e-142
#> 
#> $weights
#>      [,1]    [,2]
#> [1,]  NaN 1979597
#> [2,]  NaN 1929755
#> 
#> $stress
#> [1] 289.1415
#> 
#> $iterations
#> [1] 21
#> 
#> $n_individuals
#> [1] 2
#> 
#> $n_stimuli
#> [1] 5
#>