Function to plot the coverage of the Numerical Predictive Check

This function takes the output from the npc command in Perl Speaks NONMEM (PsN) and makes a coverage plot. A coverage plot for the NPC looks at different prediction intervals (PIs) for each data point and calculates the total number of data points in the data set lying outside of these PIs. The plot shows the relative amount of data points outside of their PI compared to the expected amount at that PI. In addition a confidence interval around these values are computed based on the simulated data.

Usage

npc.coverage(
  npc.info = "npc_results.csv",
  main = "Default",
  main.sub = NULL,
  main.sub.cex = 0.85,
  ...
)

Arguments

npc.info: The results file from the npc command in PsN. For example, npc_results.csv, or if the file is in a separate directory ./npc_dir1/npc_results.csv.
main: A string giving the plot title or NULL if none. "Default" creates a default title.
main.sub: Used for names above each plot when using multiple plots. Should be a vector c("Group 1","Group 2")
main.sub.cex: The size of the main.sub titles.
...: Other arguments passed to xpose.multiple.plot.default, xyplot and others. Please see these functions (and below) for more descriptions of what you can do.

Value

A list of plots

Additional arguments for the NPC coverage plots

Additional plot features

CI: Specifies whether confidence intervals (as lines, a shaded area or both) should be added to the plot. Allowed values are: "area", "lines", "both", or NULL.
mark.outside.data: Should the points outside the CI be marked in a different color to identify them. Allowed values are TRUE or FALSE.
abline: Should there be a line to mark the value of y=1? Possible values are TRUE, FALSE and NULL.

Line and area control. See plot, grid.polygon and xyplot for more details.

CI.area.col: Color of the area for the CI. Defaults to "blue"
CI.area.alpha: Transparency of the CI.area.col. Defaults to 0.3.
ab.lwd: The width of the abline. Default is 1.
ab.lty: Line type of the abline. Default is "dashed"
CI.upper.lty: Line type of the line at the upper edge of the CI.
CI.upper.col: Color of the line at the upper edge of the CI.
CI.upper.lwd: The line width of the line at the upper edge of the CI.
CI.lower.lty: The line type at the lower edge of the CI.
CI.lower.col: The color of the line at the lower edge of the CI.
CI.lower.lwd: The line width of the line at the lower edge of the CI.
obs.col: The color of the observed values.
obs.pch: The type of point to use for the observed values.
obs.lty: The type of line to use for the observed values.
obs.type: The combination of lines and points to use for the observed values. Default is "b" for both.
obs.cex: The size of the points to use for the observed values.
obs.lwd: The line width to use for the observed values.
out.col: The color of the observed values that lie outside of the CI. Only used if mark.outside.data=TRUE.
out.pch: The type of point to use for the observed values that lie outside of the CI. Only used if mark.outside.data = TRUE.
out.cex: The size of the points of the observed values that lie outside of the CI. Only used if mark.outside.data = TRUE.
out.lwd: The line width of the observed values that lie outside of the CI. Only used if mark.outside.data = TRUE.

Author

Andrew Hooker

Examples


if (FALSE) { # \dontrun{
library(xpose4)

npc.coverage()

## to read files in a directory different than the current working directory 
npc.file <- "./another_directory/npc_results.csv"
npc.coverage(npc.info=npc.file)
} # }