obj_func_bp: Objective function for Bounded probability

In Iraices/PrecisePvsBoundedP: A suggestion for the quantification of precise and bounded probability to express epistemic uncertainty in scientific assessments

Description

This is the objective function (a function to be optimized) The objective function is the unc_analysis_assessment_bp function where concentration_mu0, consumption_mu0 and consumption_change_vals_EKE are the parameters and the rest of the inputs arguments are fixed.

Usage

obj_func_bp(
  parameters,
  niter_ale,
  niter_epi,
  threshold,
  exposure_scenario,
  suff_stat_concentration,
  suff_stat_consumption,
  consumption_change_vals_EKE,
  consumption_change_probs_EKE,
  consumers_info_sample_size,
  concentration_v0,
  concentration_alpha0,
  concentration_beta0,
  sufficient_statistics_concentration,
  consumption_v0,
  consumption_alpha0,
  consumption_beta0,
  sufficient_statistics_consumption,
  consumption_event_alpha0,
  consumption_event_beta0,
  percentile = NULL
)

Arguments

parameters	parameters of the objective function (concentration_mu0, consumption_mu0, consumption_change_vals_EKE)
niter_ale	number of generated samples
niter_epi	number of generated parameters from the posterior distrbutions (it indicates the number of repetitions the assessment will be done)
threshold	safety threshold
exposure_scenario	a value that indicates if the assessment is done on an average consumption scenario by 'av' or on a high consumption scenario by 'perc_95'. Default is 'av'
suff_stat_concentration	a vector of sufficient statistics: sample_size, sample_mean and sample_sd corresponding to concentration. If sufficient_statistics_concentration = FALSE, then it is vector of observed data
suff_stat_consumption	a vector of sufficient statistics: sample_size, sample_mean and sample_sd corresponding to consumption. If sufficient_statistics_consumption = FALSE, then it is vector of observed data
consumption_change_vals_EKE	a vector of elicited values from experts
consumption_change_probs_EKE	a vector of elicited probabilities from experts
consumers_info_sample_size	a vector with the sample size of non_consumer_sample_size and consumer_sample_size
concentration_v0	prior hyperparameter v0 for the normal-gamma distribution corresponding to concentration
concentration_alpha0	prior hyperparameter alpha0 for the normal-gamma distribution corresponding to concentration
concentration_beta0	prior hyperparameter beta0 for the normal-gamma distribution corresponding to concentration
sufficient_statistics_concentration	logical; if TRUE, sufficient statistics (sample_size, sample_mean, sample_variance) corresponding to concentration are given as input data, otherwise sufficient_statistics_concentration is given as observed data. Default is TRUE
consumption_v0	prior hyperparameter v0 for the normal-gamma distribution corresponding to consumption
consumption_alpha0	prior hyperparameter alpha0 for the normal-gamma distribution corresponding to consumption
consumption_beta0	prior hyperparameter beta0 for the normal-gamma distribution corresponding to consumption
sufficient_statistics_consumption	logical; if TRUE, sufficient statistics (sample_size, sample_mean, sample_variance) corresponding to consumption are given as input data, otherwise sufficient_statistics_consumption is given as observed data. Default is TRUE
consumption_event_alpha0	prior hyperparameter alpha0 for the beta distribution corresponding to consumption event
consumption_event_beta0	prior hyperparameter beta0 for the beta distribution corresponding to consumption event
percentile	a value between 1 and 100 which indicates a percentile. By default is NULL

Value

expected_frequency_exceeding the expected value of the frequency of exceeding the threshold

Iraices/PrecisePvsBoundedP documentation built on Jan. 18, 2021, 11:32 p.m.