Ordered Optimal Classification for ordinal scales

Usage

morie_spatial_voting_ordered_oc(Y, n_dims = 2L, max_iter = 500L, tol = 1e-06)

Arguments

Y: Respondent-by-item ordinal response matrix.
n_dims: Latent dimensions.
max_iter: Maximum iterations.
tol: Tolerance (unused; kept for API parity).

Value

List with ideal_points, cutpoints, normals, correct_class, iterations.

References

Hare, C., Liu, T.-P., and Lupton, R. N. (2018). "What Ordered Optimal Classification reveals about ideological structure, cleavages, and polarization in the American mass public." Public Choice, 176(1), 57-78.

Examples

set.seed(1)
Y <- matrix(sample(1:4, 60, replace = TRUE), 15, 4)
morie_spatial_voting_ordered_oc(Y, n_dims = 1L, max_iter = 20L)
#> $ideal_points
#>              [,1]
#>  [1,]  1.37095845
#>  [2,] -0.56469817
#>  [3,]  0.36312841
#>  [4,]  0.63286260
#>  [5,]  0.40426832
#>  [6,] -0.10612452
#>  [7,]  1.51152200
#>  [8,] -0.09465904
#>  [9,]  2.01842371
#> [10,] -0.06271410
#> [11,]  1.30486965
#> [12,]  2.28664539
#> [13,] -1.38886070
#> [14,] -0.27878877
#> [15,] -0.13332134
#> 
#> $cutpoints
#> $cutpoints[[1]]
#> [1]  0.4342372  0.5595366 -0.9767794
#> 
#> $cutpoints[[2]]
#> [1] -0.4245716 -1.1417692
#> 
#> $cutpoints[[3]]
#> [1] -0.01524211  0.07868657 -0.96886613
#> 
#> $cutpoints[[4]]
#> [1] -0.5185655 -0.2199394 -0.3836984
#> 
#> 
#> $normals
#>      [,1]
#> [1,]    1
#> [2,]   -1
#> [3,]   -1
#> [4,]   -1
#> 
#> $correct_class
#> [1] 0.2
#> 
#> $iterations
#> [1] 2
#>