MLSMU6 alternating least-squares unfolding

Multidimensional Least-Squares Metric Unfolding (Poole 1984; Bakker & Poole 2013). Alternates between respondent and stimulus coordinates, restarting from random seeds and keeping the lowest-stress fit.

Usage

morie_spatial_voting_mlsmu6(
  D,
  n_dims = 2L,
  max_iter = 200L,
  tol = 1e-06,
  n_restarts = 5L
)

Arguments

D

Respondent-by-stimulus distance/rating matrix.

n_dims

Number of latent dimensions.

max_iter

Maximum alternations per restart.

tol

Convergence tolerance on relative stress change.

n_restarts

Number of random restarts.

Value

A list with respondent_coords, stimulus_coords, stress, iterations, converged.

References

Poole, K. T. (1984). "Least Squares Metric, Unidimensional Unfolding." Psychometrika, 49(3). Bakker, R. and Poole, K. T. (2013).

Examples

D <- matrix(stats::runif(20 * 6), 20, 6)
morie_spatial_voting_mlsmu6(D, n_dims = 2, n_restarts = 1, max_iter = 50)
#> $respondent_coords
#>             [,1]         [,2]
#>  [1,]  0.9104832 -0.032351540
#>  [2,] -0.5808314 -1.165859008
#>  [3,]  0.1223089  0.077093926
#>  [4,]  0.3524386  1.137103498
#>  [5,]  0.1623352  1.670596118
#>  [6,] -0.2385060 -0.122998796
#>  [7,]  1.0189598  0.006142552
#>  [8,] -0.2125562 -1.155509727
#>  [9,]  1.4109146  0.560035563
#> [10,] -0.2251273 -0.260078975
#> [11,]  0.8623048  0.553640053
#> [12,]  1.6255390  0.731075171
#> [13,] -1.2186330  1.007619017
#> [14,] -0.3850404 -0.245129519
#> [15,] -0.2357326  0.614528963
#> [16,]  0.3412829 -1.116651351
#> [17,] -0.3613270 -0.403336109
#> [18,] -2.1961192 -0.445549782
#> [19,] -2.0257416 -1.655870587
#> [20,]  0.8730479  0.245500534
#> 
#> $stimulus_coords
#>            [,1]        [,2]
#> [1,]  0.1183383 -0.32254720
#> [2,] -0.1046947  0.51241550
#> [3,]  0.4234505 -0.16849490
#> [4,] -0.2207089  0.22555539
#> [5,] -0.4706295  0.09006151
#> [6,]  0.2542443 -0.33699030
#> 
#> $stress
#> [1] 127.0262
#> 
#> $iterations
#> [1] 50
#> 
#> $converged
#> [1] FALSE
#>