inst/examples/bayesian_swissmetro/plot_b19_individual_level_parameters.R

In rbiogeme: Interface to 'Biogeme'

#!/usr/bin/env Rscript

# b19. Posterior means of individual-level parameters.
#
# The native example reads b05_normal_mixture.nc and calls
# BayesianResults.posterior_mean_by_observation("b_time_rnd"). To keep this R
# example runnable from a clean directory, it contains the complete b05 model
# specification and creates that prerequisite NetCDF file afresh first.

library(rbiogeme)

# prepare_swissmetro_example() is defined in ../swissmetro/example_utils.R.
# It handles --data, --python, and --output while this file keeps both the
# prerequisite model and the b19 post-estimation operation explicit.
script_path <- commandArgs(trailingOnly = FALSE)
script_path <- sub("^--file=", "", script_path[startsWith(script_path, "--file=")][[1L]])
source(file.path(dirname(normalizePath(script_path)), "..", "swissmetro", "example_utils.R"))

build_b05_normal_mixture_model <- function(database) {
  positive_lower_bound <- 1e-5
  asc_car <- biogeme_beta("asc_car", start = 0)
  asc_train <- biogeme_beta("asc_train", start = 0)
  asc_sm <- biogeme_beta("asc_sm", start = 0, fixed = TRUE)
  b_cost <- biogeme_beta("b_cost", start = 0, upper = 0)
  b_time <- biogeme_beta("b_time", start = 0, upper = 0)
  b_time_s <- biogeme_beta("b_time_s", start = 10, lower = positive_lower_bound)
  b_time_eps <- draw("b_time_eps", "NORMAL")
  b_time_rnd <- distributed_parameter(
    "b_time_rnd",
    b_time + b_time_s * b_time_eps
  )

  utilities <- list(
    `1` = asc_train + b_time_rnd * variable("TRAIN_TT_SCALED") +
      b_cost * variable("TRAIN_COST_SCALED"),
    `2` = asc_sm + b_time_rnd * variable("SM_TT_SCALED") +
      b_cost * variable("SM_COST_SCALED"),
    `3` = asc_car + b_time_rnd * variable("CAR_TT_SCALED") +
      b_cost * variable("CAR_CO_SCALED")
  )
  availability <- list(
    `1` = variable("TRAIN_AV_SP"),
    `2` = variable("SM_AV"),
    `3` = variable("CAR_AV_SP")
  )
  biogeme_model(
    database = database,
    formula = logit_log_probability(
      utilities = utilities,
      availability = availability,
      alternative = variable("CHOICE")
    ),
    control = biogeme_control(
    output_directory = prepared$output,
      model_name = "b05_normal_mixture",
      generate_html = TRUE,
      generate_yaml = TRUE,
      generate_netcdf = TRUE
    )
  )
}

prepared <- prepare_swissmetro_example(
  commandArgs(trailingOnly = TRUE),
  default_model = "b19_individual_level_parameters"
)

# The native b19 input is b05_normal_mixture.nc. Delete all exact b05 outputs
# first so this prerequisite never reuses stale Bayesian results.
unlink(file.path(prepared$output, c(
  "b05_normal_mixture.yaml",
  "b05_normal_mixture.nc",
  "b05_normal_mixture.html",
  "__b05_normal_mixture.iter"
)), force = TRUE)

database <- swissmetro_data(prepared$data)
model <- build_b05_normal_mixture_model(database)
fit <- bayesian_estimate(
  model,
  model_name = "b05_normal_mixture",
  control = model$control
)

print(summary(fit))
print(bayesian_stored_variables(fit))

# This delegates NetCDF loading and the posterior reduction to native
# BayesianResults. The returned ordinary data frame has one row per native
# observation coordinate and a column named b_time_rnd.
individual_parameters <- bayesian_posterior_mean_by_observation(
  fit,
  variable_name = "b_time_rnd"
)
print(individual_parameters)
invisible(individual_parameters)