add.model.comp | R Documentation |
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.
add.model.comp(
object,
object.ref = NULL,
onlyfirst = FALSE,
inclZeroWRES = FALSE,
subset = xsubset(object),
main = "Default",
force.wres = FALSE,
...
)
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 |
main |
The title of the plot. If |
force.wres |
Should we use the WRES in the plots instead of CWRES
(logical |
... |
Other arguments passed to |
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.
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.
E. Niclas Jonsson, Mats Karlsson, Andrew Hooker & Justin Wilkins
xpose.plot.default
,
xpose.panel.default
, xyplot
,
compute.cwres
, xpose.prefs-class
,
xpose.data-class
## Not run:
## 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)
## End(Not run)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.