Model residuals plotted against the independent variable (IDV).
The residuals can be one of:
RES: model residuals
WRES: weighted model residuals
CWRES: conditional weighted model residuals
EWRES/ECWRES: Monte Carlo based model residuals
NPDE: Normalized prediction distribution error
res_vs_idv(
xpdb,
mapping = NULL,
res = "CWRES",
group = "ID",
type = "pls",
title = "@y vs. @x | @run",
subtitle = "Ofv: @ofv",
caption = "@dir",
tag = NULL,
log = NULL,
guide = TRUE,
facets,
.problem,
quiet,
...
)
absval_res_vs_idv(
xpdb,
mapping = NULL,
res = "CWRES",
group = "ID",
type = "pls",
title = "@y vs. @x | @run",
subtitle = "Ofv: @ofv",
caption = "@dir",
tag = NULL,
log = NULL,
guide = FALSE,
facets,
.problem,
quiet,
...
)
An xpose database object.
List of aesthetics mappings to be used for the xpose plot
(e.g. point_color
).
Type of residual to be used. Default is "CWRES".
Grouping variable to be used for lines.
String setting the type of plot to be used. Can be points 'p', line 'l', smooth 's' and text 't' or any combination of the four.
Plot title. Use NULL
to remove.
Plot subtitle. Use NULL
to remove.
Page caption. Use NULL
to remove.
Plot identification tag. Use NULL
to remove.
String assigning logarithmic scale to axes, can be either '', 'x', y' or 'xy'.
Enable guide display (e.g. unity line).
Either a character string to use facet_wrap_paginate
or a formula to use facet_grid_paginate
.
The $problem number to be used. By default returns the last estimation problem.
Logical, if FALSE
messages are printed to the console.
Any additional aesthetics to be passed on xplot_scatter
.
Plots can be customized by mapping arguments to specific layers. The naming convention is layer_option where layer is one of the names defined in the list below and option is any option supported by this layer e.g. point_color = 'blue', smooth_method = 'lm', etc.
point: options to geom_point
line: options to geom_line
guide: options to geom_abline
smooth: options to geom_smooth
text: options to geom_text
xscale: options to scale_x_continuous
or scale_x_log10
yscale: options to scale_y_continuous
or scale_y_log10
Template titles can be used to create highly informative diagnostics plots.
They can be applied to any plot title, subtitle, caption and tag. Template titles
are defined via a single string containing key variables staring with a `@` (e.g. `@ofv`)
which will be replaced by their actual value when rendering the plot.
For example `'@run, @nobs observations in @nind subjects'` would become
`'run001, 1022 observations in 74 subjects'`. The available key variables
are listed under template_titles
.
# Standard residual
res_vs_idv(xpdb_ex_pk, res = c('IWRES', 'CWRES'))
#> Using data from $prob no.1
#> Filtering data by EVID == 0
#> Tidying data by ID, SEX, MED1, MED2, DOSE ... and 24 more variables
#> `geom_smooth()` using formula = 'y ~ x'
#> `geom_smooth()` using formula = 'y ~ x'
# Absolute value of the residuals
absval_res_vs_idv(xpdb_ex_pk, res = 'CWRES')
#> Using data from $prob no.1
#> Filtering data by EVID == 0
#> `geom_smooth()` using formula = 'y ~ x'
#> `geom_smooth()` using formula = 'y ~ x'