pcSobol-class: An S4 class to return the results of sensitivity analyses

In JohnNay/sa: Sensitivity Analysis for Complex Computational Models

Description

An S4 class to return the results of sensitivity analyses

Plots pcSobol S4 object

Usage

## S4 method for signature 'pcSobol'
print(x, ...)

## S4 method for signature 'pcSobol'
show(object)

## S4 method for signature 'pcSobol'
plot(x, outcome_var = "Outcome",
  legend_pos = c("topright", "bottomright", "bottomleft", "topleft"),
  ptype = c("all", "fo", "total"))

Arguments

x	The result slot of an object created by sobol_sa.
...	ignored
object	S4 pcSA object
outcome_var	Optional character vector for labeling the outcome variable in the plot. Default is "Outcome".
legend_pos	Character vector that sets the position of the legend to one of: "topright", "bottomright", "bottomleft", or "topleft".
ptype	Optional Character vector for first order, total, or all effects c("all", "fo", "total").

Value

Returns a ggplot2 plot.

Methods (by generic)

• print: An S4 method for printing a pcSobol S4 object

• show: An S4 method for showing a pcSobol S4 object

• plot: Plot First-Order (Main) Effects and Total Effects from a Sobol Sensitivity Analysis of a Simulation Model in the Same Plot

  This is function of the eat package. sobol_sa conducts a global variance decomposition, and then this can be used to plot it.

Slots

call

Language from the call of the function.

result

sobol2007 s3 class from sensitivity package.

timing

Numeric vector length one with the total elapsed time it took to execute.

session

the results from calling sessionInfo() at end of pc_sa function.

References

J. C. Thiele, W. Kurth, V. Grimm, Facilitating Parameter Estimation and Sensitivity Analysis of Agent-Based Models: A Cookbook Using NetLogo and R. Journal of Artificial Societies and Social Simulation. 17, 11 (2014).

Examples

fake_abm <- function(params, out) {
  x1 <- params[1]
  x2 <- params[2]
  if (out=="sq") return(x1^2 + x2 + rnorm(1, 0))
  if (out=="ident") return(x1 + x2 + rnorm(1, 0))
}
inputs <- lapply(list(param1 = NA, param2 = NA), 
                 function(x) list(random_function = "qunif",
                                  ARGS = list(min = 0, max = 1)))
s <- sobol_sa(fake_abm, inputs, "sq")
plot(s)

JohnNay/sa documentation built on May 7, 2019, 12:01 p.m.