Skip to contents

Define probit model parameter

Source: R/RprobitB_parameter.R
RprobitB_parameter.Rd

This function creates an object of class RprobitB_parameter, which contains the parameters of a probit model.

If sample = TRUE, missing parameters are sampled. All parameters are checked against the values of P_f, P_r, J, and N.

Note that parameters are automatically ordered with respect to a non-ascending s for class identifiability.

Usage

RprobitB_parameter(
  P_f,
  P_r,
  J,
  N,
  C = 1,
  ordered = FALSE,
  alpha = NULL,
  s = NULL,
  b = NULL,
  Omega = NULL,
  Sigma = NULL,
  Sigma_full = NULL,
  beta = NULL,
  z = NULL,
  d = NULL,
  sample = TRUE
)

# S3 method for class 'RprobitB_parameter'
print(x, ..., digits = 4)

Arguments

P_f

[integer(1)]
The number of covariates connected to a fixed coefficient.

P_r

[integer(2)]
The number of covariates connected to a random coefficient.

J

[integer(1)]
The number >= 2 of choice alternatives.

N

[integer(1)]
The number of decision makers.

C

[integer(1)]
The number (greater or equal 1) of latent classes of decision makers.

ordered

[logical(1)]
If TRUE, the choice set alternatives is assumed to be ordered from worst to best.

alpha

[numeric(P_f)]
The fixed coefficient vector.

s

[numeric(C)]
The vector of class weights.

b

[matrix(nrow = P_r, ncol = C)]
The matrix of class means as columns.

Omega

[matrix(nrow = P_r * P_r, ncol = C)]
The matrix of vectorized class covariance matrices as columns.

Sigma

[matrix(nrow = J - 1, ncol = J - 1) | numeric(1)]
The differenced (wrt. alternative J) error covariance matrix.

In case of ordered = TRUE, the single error variance.

Sigma_full

[matrix(nrow = J, ncol = J)]
The error covariance matrix.

Ignored if Sigma is specified or ordered = TRUE.

Internally, Sigma_full gets differenced wrt. alternative J.

beta

[matrix(nrow = P_r, ncol = N)]
The matrix of the decider-specific coefficient vectors.

z

[numeric(N)]
The decider class allocations.

d

[numeric(J - 2)]
The logarithmic increases of the utility thresholds in the ordered probit case (ordered = TRUE).

sample

[logical(1)]
Sample missing parameters?

x

An RprobitB_parameter object.

...

[character()]
Names of parameters to be printed. If not specified, all parameters are printed.

digits

[integer(1)]
The number of decimal places.

Value

An object of class RprobitB_parameter, which is a named list with the model parameters.

Examples

RprobitB_parameter(P_f = 1, P_r = 2, J = 3, N = 10, C = 2)
#> alpha : -0.9
#> 
#> C : 2
#> 
#> s : double vector of length 2 
#> 0.76 0.24
#> 
#> b : 2 x 2 matrix of doubles 
#>      [,1] [,2]
#> [1,] -0.8  1.3
#> [2,]  3.0 -2.5
#> 
#> 
#> Omega : 4 x 2 matrix of doubles 
#>      [,1]  [,2]
#> [1,] 1.15  1.26
#> [2,] 0.06 -0.21
#> [3,] 0.06 -0.21
#> [4,]  1.2  1.98
#> 
#> 
#> Sigma : 2 x 2 matrix of doubles 
#>      [,1] [,2]
#> [1,] 2.80 1.42
#> [2,] 1.42 2.36
#> 
#> 
#> Sigma_full : 3 x 3 matrix of doubles 
#>      [,1] [,2] [,3]
#> [1,] 1.77 0.36 0.06
#> [2,] 0.36 1.27 0.03
#> [3,] 0.06 0.03 1.14
#> 
#> 
#> beta : 2 x 10 matrix of doubles 
#>       [,1]  [,2] [,3] ... [,10]
#> [1,]  1.56  0.01 0.06 ...  2.68
#> [2,] -3.14 -3.16 1.98 ...  -2.9
#> 
#> 
#> z : double vector of length 10 
#> 2 2 1 ... 2
#> 
#> d : NA
#>