RprobitB1.1.4 package

Re-label alternative specific covariates

Check model formula

Check prior parameters

Compute choice probabilities

Classify deciders preference-based

Extract model effects

Compute probit choice probabilities

Compute choice probabilities at posterior samples

Extract estimated covariance matrix of mixing distribution

Create labels for alpha

Create labels for b

Create labels for d

Create labels for Omega

Create labels for s

Create labels for Sigma

Create lagged choice covariates

Transform threshold increments to thresholds

Density of multivariate normal distribution

Sample from prior distributions

Euclidean distance

Filter Gibbs samples

Fit probit model to choice data

Extract covariates of choice occasion

Markov chain Monte Carlo simulation for the probit model

Log-likelihood in the ordered probit model

Handle missing covariates

Approximate marginal model likelihood

Compare fitted models

Extract number of model parameters

Print effect overview

Create parameters labels

Visualize choice data

Visualize fitted probit model

Autocorrelation plot of Gibbs samples

Plot class allocation (for P_r = 2 only)

Visualizing the number of classes during Gibbs sampling

Plot bivariate contour of mixing distributions

Plot marginal mixing distributions

Plot ROC curve

Visualizing the trace of Gibbs samples.

Compute point estimates

Parameter sets from posterior samples

Compute prediction accuracy

Predict choices

Check for flip in preferences after change in model scale.

Prepare choice data for estimation

Compute Gelman-Rubin statistic

Draw from Dirichlet distribution

Draw from multivariate normal distribution

RprobitB: Bayesian Probit Choice Modeling

Create object of class RprobitB_data

Create object of class RprobitB_fit

Create object of class RprobitB_gibbs_samples_statistics

Create object of class RprobitB_latent_classes

Create object of class RprobitB_normalization

Define probit model parameter

Draw from one-sided truncated normal

Draw from two-sided truncated normal

Draw from Wishart distribution

Set initial values for the Gibbs sampler

Simulate choice data

Compute sufficient statistics

Split choice data in train and test subset

Transform fitted probit model

Transformation of Gibbs samples

Transformation of parameter values

Transform differenced to non-differenced error term covariance matrix

Update and re-fit probit model

Update class means

Dirichlet process-based update of latent classes

Weight-based update of latent classes

Update utility threshold increments

Update class sizes

Update class covariances

Update coefficient vector of multiple linear regression

Update class weight vector

Update error term covariance matrix of multiple linear regression

Update latent utility vector

Update latent utility vector in the ranked probit case

Update class allocation vector

Compute WAIC value