tests/testthat/_snaps/model_fitting.md

In RprobitB: Bayesian Probit Choice Modeling

setting prior parameter works

Code
  prior
Output
  $mu_alpha_0
  [1] 0

  $Sigma_alpha_0
       [,1]
  [1,]   10

  $delta
  [1] 1

  $mu_b_0
  [1] 0 0

  $Sigma_b_0
       [,1] [,2]
  [1,]   10    0
  [2,]    0   10

  $n_Omega_0
  [1] 4

  $V_Omega_0
       [,1] [,2]
  [1,]    1    0
  [2,]    0    1

  $n_Sigma_0
  [1] 4

  $V_Sigma_0
       [,1] [,2]
  [1,]    1    0
  [2,]    0    1

  $mu_d_0
  [1] NA

  $Sigma_d_0
  [1] NA

  attr(,"class")
  [1] "RprobitB_prior" "list"

RprobitB_latent_class setting works

Code
  RprobitB_latent_classes(list(C = 2))
Output
  Latent classes
  C = 2

Code
  (out <- RprobitB_latent_classes(list(wb_update = TRUE, dp_update = TRUE)))
Output
  Latent classes
  Dirichlet process update: TRUE 
  Weight-based update: TRUE 
  Maximum classes: 10 
  Updating buffer: 50 
  Minimum class weight: 0.01 
  Maximum class weight: 0.7 
  Mimumum class distance: 0.1

Code
  str(out)
Output
  List of 10
   $ wb_update   : logi TRUE
   $ dp_update   : logi TRUE
   $ C           : num 1
   $ Cmax        : num 10
   $ buffer      : num 50
   $ epsmin      : num 0.01
   $ epsmax      : num 0.7
   $ deltamin    : num 0.1
   $ deltashift  : num 0.5
   $ class_update: logi TRUE
   - attr(*, "class")= chr "RprobitB_latent_classes"

building of RprobitB_normalization works

Code
  RprobitB_normalization(level = "B", scale = "price := -1", form = form, re = re,
    alternatives = alternatives, base = "B")
Output
  Level: Utility differences with respect to alternative 'B'.
  Scale: Coefficient of effect 'price' (alpha_1) fixed to -1.

Gibbs sampling works

Code
  print(model)
Output
  Probit model 'choice ~ a | b | c'.

Code
  summary(model)
Output
  Probit model
  Formula: choice ~ a | b | c 
  R: 2000, B: 1000, Q: 1
  Level: Utility differences with respect to alternative 'B'.
  Scale: Coefficient of the 1. error term variance fixed to 1.

  Gibbs sample statistics
            true    mean      sd      R^
   alpha

       1   -1.09   -1.11    0.06    1.00
       2    1.52    1.51    0.09    1.01
       3   -0.61   -0.62    0.06    1.01
       4   -0.12   -0.08    0.05    1.03
       5    0.49    0.37    0.05    1.00

   Sigma

     1,1    1.00    1.00    0.00    1.00

Code
  print(coef(model))
Output
           Estimate   (sd)
  1     a     -1.11 (0.06)
  2   b_A      1.51 (0.09)
  3 ASC_A     -0.62 (0.06)
  4   c_A     -0.08 (0.05)
  5   c_B      0.37 (0.05)

Ordered probit model estimation works

Code
  print(model)
Output
  Probit model 'opinion_on_sth ~ age + gender'.

Code
  summary(model)
Output
  Probit model
  Formula: opinion_on_sth ~ age + gender 
  R: 1000, B: 500, Q: 1
  Level: Fixed first utility threshold to 0.
  Scale: Error term variance fixed to 1.

  Gibbs sample statistics
            true    mean      sd      R^
   alpha

       1    0.59    0.64    0.04    1.01
       2    1.18    1.17    0.07    1.00

   Sigma

     1,1    1.00    1.00    0.00      NA

   d

       1    0.00   -0.55    0.07    1.00
       2    1.00    0.54    0.04    1.00
       3    2.00    2.78    0.22    1.05

Code
  print(coef(model))
Output
            Estimate   (sd)
  1    age      0.64 (0.04)
  2 gender      1.17 (0.07)

Ranked probit model estimation works

Code
  print(model)
Output
  Probit model 'product ~ price'.

Code
  summary(model)
Output
  Probit model
  Formula: product ~ price 
  R: 1000, B: 500, Q: 1
  Level: Utility differences with respect to alternative 'C'.
  Scale: Coefficient of the 1. error term variance fixed to 1.

  Gibbs sample statistics
            true    mean      sd      R^
   alpha

       1    1.00    1.04    0.05    1.00
       2   -0.56   -0.57    0.05    1.00
       3   -1.63   -1.65    0.09    1.01

   Sigma

     1,1    1.00    1.00    0.00    1.00
     1,2    0.39    0.39    0.08    1.03
     2,2    0.84    0.95    0.13    1.00

Code
  print(coef(model))
Output
           Estimate   (sd)
  1 price      1.04 (0.05)
  2 ASC_A     -0.57 (0.05)
  3 ASC_B     -1.65 (0.09)

RprobitB documentation built on Aug. 26, 2025, 1:08 a.m.