generate_pa_inputs: Generate probabilistic model inputs.

View source: R/fct_hemodel.R

generate_pa_inputsR Documentation

Generate probabilistic model inputs.

Description

This function generates the probabilistic model inputs for the example health economic model developed to test the functionalities of the package.

Usage

generate_pa_inputs(n_sim = 10000, sd_var = 0.2, seed_num = 452)

Arguments

n_sim

integer. Number of probabilistic value to draw for each model input. Default is 10,000.

sd_var

numeric. Determines the standard error of the mean to use for the normal distributions when the standard error not known. Default is 0.2 (20%).

seed_num

integer. The seed number to use when drawing the probabilistic values. Default is 452.

Value

A dataframe. A description of the variables of the returned dataframe is available in the documentation of the df_pa dataframe.

Examples

# Generating deterministic model inputs and storing them in an object.
df_inputs_prob <- generate_pa_inputs()

Xa4P/pacheck documentation built on April 14, 2025, 1:51 p.m.