This creates a stack of four plots, comparing absolute values of PRED, absolute values of IPRED, delta CWRES (or WRES) and delta IWRES estimates for the two specified model fits.
Usage
add.model.comp(
object,
object.ref = NULL,
onlyfirst = FALSE,
inclZeroWRES = FALSE,
subset = xsubset(object),
main = "Default",
force.wres = FALSE,
...
)
Arguments
- object
An xpose.data object.
- object.ref
An xpose.data object. If not supplied, the user will be prompted.
- onlyfirst
Logical value indicating whether only the first row per individual is included in the plot.
- inclZeroWRES
Logical value indicating whether rows with WRES=0 is included in the plot. The default is TRUE.
- subset
A string giving the subset expression to be applied to the data before plotting. See
xsubset
.- main
The title of the plot. If
"Default"
then a default title is plotted. Otherwise the value should be a string like"my title"
orNULL
for no plot title.- force.wres
Should we use the WRES in the plots instead of CWRES (logical
TRUE
orFALSE
)- ...
Other arguments passed to
link{xpose.plot.default}
.
Value
Returns a stack of plots comprising comparisons of absolute values of PRED, absolute values of IPRED, absolute differences in CWRES (or WRES) and absolute differences in IWRES for the two specified runs.
Details
Four model comparison plots are displayed in sequence.
Conditional weighted residuals (CWRES) require some extra steps to
calculate. See compute.cwres
for details.
A wide array of extra options controlling xyplots are available. See
xpose.plot.default
for details.
Examples
if (FALSE) {
## We expect to find the required NONMEM run and table files for runs
## 5 and 6 in the current working directory
xpdb5 <- xpose.data(5)
xpdb6 <- xpose.data(6)
## A vanilla plot, without prompts
add.model.comp(xpdb5, xpdb6, prompt = FALSE)
## Custom colours and symbols, no user IDs
add.model.comp(xpdb5, xpdb6, cex=0.6, pch=8, col=1, ids=NULL)
}