plots: Uncertainty and Sensitivity Plots.

Description Usage Arguments Details Examples

Description

The functions listed here are used in uncertainty and sensitivity estimation.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
plotcv(obj, stack = FALSE, index.res = 1:get.noutputs(obj),
  col = index.res, quant = 0.99, ...)

plotecdf(obj, stack = FALSE, index.res = 1:get.noutputs(obj),
  col = index.res, xlab = NULL, ...)

plotprcc(obj, index.res = 1:dim(obj$res)[2], col = "orange", ylab = NULL,
  ...)

plotscatter(obj, res = NULL, index.data = NULL, index.res = NULL,
  add.lm = TRUE, ylab = NULL, ...)

Arguments

obj

The LHS or PLUE object containing the simulation results to be plotted.

NOTICE: plotecdf and plotcv only accept LHS objects! For plotting the likelihood profile from a PLUE object, simply use plot(obj)

stack

If the results is a data.frame with several variables, stack=FALSE generates a series of plots, and stack=TRUE generates a single plot with the ECDF from all variables identified by different colors.

index.res

An optional vector indicating which columns from the results are to be plotted.

col

An optional vector indicating the colors to be used.

quant

Maximum quantile to be plotted on the ecdf (used to cut off extreme values in the labels)

...

Additional parameters to be passed to the lower level plotting function.

xlab, ylab

Labels for the x axis (ecdf) or y axis(prcc). The functions use the name provided in the res.names argument from the LHS function if left blank.

res

A data.frame consisting of the model results to be plotted on the y axis, if 'obj' is passed as a data.frame. If 'obj' is an LHS/PLUE object, this parameter is ignored.

index.data

An optional vector with the indices of the data columns to be plotted.

add.lm

Boolean. Whether to include a simple linear model on the plots. Defaults to TRUE.

Details

The function plotscatter produces a series of scatterplots from data.

The function plotecdf plots the empirical cumulative density function from an LHS object or PLUE object.

The function plotprcc plots the partial rank correlation coefficient from an LHS object or PLUE object.

Finally, the plotcv function plots the empirical cummulative density function (ecdf) of the coefficient of variation of the LHS resulting from a stochastic simulation, along with a dotted line representing the coefficient of variation of the whole result set. See the 'multiple' vignette for examples and interpretation.

The function plotscatter accepts an alternative invocation of plotscatter(obj, res) in which obj is a data.frame consisting on the data to be plotted on the x axis, and res is a data.frame consisting on the model results to be plotted on the y axis.

Examples

1
2
3
4
myLHS <- LHS(model=function(x) x[,1]+x[,2]*x[,3], factors=3, N=20, res.names="My Output")
plotecdf(myLHS, main="ECDF plot")
plotprcc(myLHS, main="PRCC plot")
plotscatter(myLHS)

Example output

Loading required package: Hmisc
Loading required package: lattice
Loading required package: survival
Loading required package: Formula
Loading required package: ggplot2

Attaching package: 'Hmisc'

The following objects are masked from 'package:base':

    format.pval, round.POSIXt, trunc.POSIXt, units

pse documentation built on May 2, 2019, 12:56 a.m.