run_owsa_det: Run deterministic one-way sensitivity analysis (OWSA)

In DARTH-git/dampack: Decision-Analytic Modeling Package

Run deterministic one-way sensitivity analysis (OWSA)

Description

This function runs a deterministic one-way sensitivity analysis (OWSA) on a given function that produces outcomes.

Usage

run_owsa_det(
  params_range,
  params_basecase,
  nsamp = 100,
  FUN,
  outcomes = NULL,
  strategies = NULL,
  progress = TRUE,
  ...
)

Arguments

params_range	data.frame with 3 columns in the following order: "pars", "min", and "max". The number of samples from this range is determined by nsamp. "pars" are the parameters of interest and must be a subset of the parameters from params_basecase.
params_basecase	a named list of basecase values for input parameters needed by FUN, the user-defined function.
nsamp	number of sets of parameter values to be generated. If NULL, 100 parameter values are used
FUN	function that takes the basecase in params_basecase and ... to produce the outcome of interest. The FUN must return a dataframe where the first column are the strategy names and the rest of the columns must be outcomes.
outcomes	string vector with the outcomes of interest from FUN produced by nsamp
strategies	vector of strategy names. The default NULL will use strategy names in FUN
progress	TRUE or FALSE for whether or not function progress should be displayed in console.
...	Additional arguments to user-defined FUN

Value

A list containing dataframes with the results of the sensitivity analyses. The list will contain a dataframe for each outcome specified. List elements can be visualized with plot.owsa, owsa_opt_strat and owsa_tornado from dampack

Details

• params_range

  • "pars" are the names of the input parameters of interest. These are the parameters that will be varied in the deterministic sensitivity analysis. variables in "pars" column must be a subset of variables in params_basecase

  • "min" and "max" are the mininum and maximum values of the parameters of interest.

DARTH-git/dampack documentation built on Sept. 25, 2024, 4:48 p.m.