cdf_plot: Plot a cumulative distribution function (CDF)

Description

This function creates a CDF plot. Input data for the plots is provided by a data frame with the same structure as the "CDF" output from cont_analysis. Confidence limits for the CDF also are plotted.

Usage

cdf_plot(
  cdfest,
  var = NULL,
  subpop = NULL,
  subpop_level = NULL,
  units_cdf = "Percent",
  type_cdf = "Continuous",
  log = "",
  xlab = NULL,
  ylab = NULL,
  ylab_r = NULL,
  main = NULL,
  legloc = NULL,
  confcut = 0,
  conflev = 95,
  cex.main = 1.2,
  cex.legend = 1,
  ...
)

Value

A plot of a variable's CDF estimates associated confidence limits.

Arguments

cdfest: Data frame with the same structure as the "CDF" output from cont_analysis.
var: If cdfest has multiple variables in the "Indicator" column, then var is the single variable to be plotted. The default is NULL, which assumes that only one variable is in the "Indicator" column of cdfest.
subpop: If cdfest has multiple variables in the "Type" column, then subpop is the single variable to be plotted. The default is NULL, which assumes that only one variable is in the "Type" column of cdfest.
subpop_level: If cdfest has multiple levels of subpop in the "Subpopulation" column, then subpop_level is the single level to be plotted. The default is NULL, which assumes that only one level is in the "Subpopulation" column of cdfest.
units_cdf: Indicator for the label utilized for the left side y-axis and the values used for the left side y-axis tick marks, where "Percent" means the label and values are in terms of percent of the population, and "Units" means the label and values are in terms of units (count, length, or area) of the population. The default is "Percent".
type_cdf: Character string consisting of the value "Continuous" or "Ordinal" that controls the type of CDF plot. The default is "Continuous".
log: Character string consisting of the value "" or "x" that controls whether the x axis uses the original scale ("") or the base 10 logarithmic scale ("x"). The default is "".
xlab: Character string providing the x-axis label. If this argument equals NULL, then the indicator name is used as the label. The default is NULL.
ylab: Character string providing the left side y-axis label. If argument units_cdf equals "Units", a value should be provided for this argument. Otherwise, the label will be "Percent". The default is "Percent".
ylab_r: Character string providing the label for the right side y-axis (and, hence, determining the values used for the right side y-axis tick marks), where NULL means a right side y-axis is not created. If this argument equals "Same", the right side y-axis will have the same label and tick mark values as the left side y-axis. If this argument equals a character string other than "Same", the right side y-axis label will be the value provided for argument ylab_r, and the right side y-axis tick mark values will be determined by the choice not utilized for argument units_cdf, which means that the default value of argument units_cdf (i.e., "Percent") will result in the right side y-axis tick mark values being expressed in terms of units of the population (i.e., count, length, or area). The default is NULL.
main: Character string providing the plot title. The default is NULL.
legloc: Indicator for location of the plot legend, where "BR" means bottom right, "BL" means bottom left, "TR" means top right, "TL" means top left, and NULL means no legend. The default is NULL.
confcut: Numeric value that controls plotting confidence limits at the CDF extremes. Confidence limits for CDF values (percent scale) less than confcut or greater than 100 minus confcut are not plotted. A value of zero means confidence limits are plotted for the complete range of the CDF. The default is 0.
conflev: Numeric value of the confidence level used for confidence limits. The default is 95.
cex.main: Expansion factor for the plot title. The default is 1.2.
cex.legend: Expansion factor for the legend title. The default is 1.
...: Additional arguments passed to the plot.default function (aside from those already used and ylim).

Author

Tom Kincaid Kincaid.Tom@epa.gov

Examples

Run this code

if (FALSE) {
dframe <- data.frame(
  siteID = paste0("Site", 1:100),
  wgt = runif(100, 10, 100),
  xcoord = runif(100),
  ycoord = runif(100),
  stratum = rep(c("Stratum1", "Stratum2"), 50),
  ContVar = rnorm(100, 10, 1),
  All_Sites = rep("All Sites", 100),
  Resource_Class = rep(c("Good", "Poor"), c(55, 45))
)
myvars <- c("ContVar")
mysubpops <- c("All_Sites", "Resource_Class")
mypopsize <- data.frame(
  Resource_Class = c("Good", "Poor"),
  Total = c(4000, 1500)
)
myanalysis <- cont_analysis(dframe,
  vars = myvars, subpops = mysubpops,
  siteID = "siteID", weight = "wgt", xcoord = "xcoord", ycoord = "ycoord",
  stratumID = "stratum", popsize = mypopsize
)
keep <- with(myanalysis$CDF, Type == "Resource_Class" &
  Subpopulation == "Good")
par(mfrow = c(2, 1))
cdf_plot(myanalysis$CDF[keep, ],
  xlab = "ContVar",
  ylab = "Percent of Stream Length", ylab_r = "Stream Length (km)",
  main = "Estimates for Resource Class: Good"
)
cdf_plot(myanalysis$CDF[keep, ],
  xlab = "ContVar",
  ylab = "Percent of Stream Length", ylab_r = "Same",
  main = "Estimates for Resource Class: Good"
)
}