confPlot: Plot confidence intervals of repeated estimates for one...

In andrew-edwards/sizeSpectra: Fitting Size Spectra to Ecological Data Using Maximum Likelihood

Plot confidence intervals of repeated estimates for one method

Description

Plot confidence intervals of the repeated estimates of b for one method. Gets called eight times to produce Figure 4 of MEE paper and Figure 5 of MEPS paper. Plots horizontal lines for the intervals, colour coded as to whether or not they include the true value of b.

Usage

confPlot(
  repConf,
  legName,
  b.true = b.known,
  inCol = "darkgrey",
  outCol = "blue",
  vertCol = "red",
  pchVal = 20,
  cexVal = 0,
  xLim = NULL,
  colourCode = TRUE,
  vertThick = 1,
  horizThick = 0.3,
  thin = 33,
  horizLines = FALSE,
  horizLinesOut = TRUE,
  horizLinesIn = TRUE,
  yLab = "Sample number",
  yTicks = seq(0, 300, 50),
  yLabels = TRUE,
  vertFirst = FALSE,
  insetVal = c(-0.08, -0.06),
  insetVal2 = c(-0.08, 0.07),
  xsmallticks = NULL,
  legLoc = "topleft"
)

Arguments

repConf	data frame with columns confMin and confMax, the minimum and maximum of the 95\ corresponding to each simulated data set. When confPlot is called, for some methods b = slope - 1 or b = slope - 2, which gets done in the call.
legName	legend name for that panel
b.true	true known value of b (as used for simulating the data)
inCol	colour for intervals that b.true is inside
outCol	colour for intervals that b.true is outside
vertCol	colour of vertical line for b.true
pchVal	pch for points at endpoints of intervals
cexVal	size of points (default of 0 does not plot them)
xLim	x-axis range
colourCode	colour code the figure
vertThick	thickness of vertical line for b.true
horizThick	thickness of horizontal lines
thin	number of values to thin the values for plotting. Only works for 33 or 99 for now (since they are factors of 9999, the value used in simulations).
horizLines	whether to plot horizontal grey lines or not as example (not needed now since doing lines for intervals)
horizLinesOut	whether to plot horizontal lines for intervals for which true value is outside the interval
horizLinesIn	whether to plot horizontal lines for intervals for which true value is inside the interval
yLab	label for y-axis
yTicks	where to have tickmarks on y-axis
yLabels	whether or not to label tickmarks on y-axis
vertFirst	whether or not to plot vertical line first in order to see the horizontal lines better (for LCD plot)
insetVal	inset shift for naming the panel
insetVal2	inset shift for printing observed coverage percentage
xsmallticks	where to put unlabelled small tick marks on x-axis
legLoc	where to put the legend, as the first argument in legend()

Value

plots a panel and returns what is plotted as a sorted data frame

Author(s)

Andrew Edwards

andrew-edwards/sizeSpectra documentation built on June 28, 2023, 7:09 p.m.