prepares data for choice modelling with mlogit, and estimates a basic choice model.

Defines indices, makes changes to variable formats and data structure to prepare for choice modelling with mlogit.

Usage

Repeatr_4(mydf = NULL, output_dir = NULL)

Arguments

mydf: optional dataframe to be used. If omitted, the default dataframe will be used. Example of use: ml_Repeatr4 <- Repeatr_4()
output_dir: Optional directory to save the rebuilt data/results_ml_Repeatr4.rda/data/vcovmat_ml_Repeatr4.rda into. If omitted, defaults to data/ under the current working directory.

Value

A data frame (results_ml_Repeatr4) of the mlogit coefficient table: one row per model covariate, with columns for the estimate, standard error, z-value and p-value. Also saved to data/results_ml_Repeatr4.rda, alongside the corresponding vcovmat_ml_Repeatr4.rda (needed by diffr/rankr) - both are always written together by this function so they can never go out of sync with each other.

Examples

if (FALSE) { # \dontrun{
# Fits a real mlogit choice model on the full dataset - the slow step
# in the pipeline (see vignette("Rebuilding-the-Data")), not run here.
results_ml_Repeatr4 <- Repeatr_4(output_dir = tempdir())
} # }