plot.vsel  R Documentation 
This is the plot()
method for vsel
objects (returned by varsel()
or
cv_varsel()
). It visualizes the predictive performance of the reference
model (possibly also that of some other "baseline" model) and that of the
submodels along the fulldata predictor ranking. Basic information about the
(CV) variability in the ranking of the predictors is included as well (if
available; inferred from cv_proportions()
). For a tabular representation,
see summary.vsel()
.
## S3 method for class 'vsel'
plot(
x,
nterms_max = NULL,
stats = "elpd",
deltas = FALSE,
alpha = 2 * pnorm(1),
baseline = if (!inherits(x$refmodel, "datafit")) "ref" else "best",
thres_elpd = NA,
resp_oscale = TRUE,
point_size = 3,
bar_thickness = 1,
ranking_nterms_max = NULL,
ranking_abbreviate = FALSE,
ranking_abbreviate_args = list(),
ranking_repel = NULL,
ranking_repel_args = list(),
ranking_colored = FALSE,
cumulate = FALSE,
text_angle = NULL,
...
)
x 
An object of class 
nterms_max 
Maximum submodel size (number of predictor terms) for which
the performance statistics are calculated. Using 
stats 
One or more character strings determining which performance
statistics (i.e., utilities or losses) to estimate based on the
observations in the evaluation (or "test") set (in case of
crossvalidation, these are all observations because they are partitioned
into multiple test sets; in case of

deltas 
If 
alpha 
A number determining the (nominal) coverage 
baseline 
For 
thres_elpd 
Only relevant if 
resp_oscale 
Only relevant for the latent projection. A single logical
value indicating whether to calculate the performance statistics on the
original response scale ( 
point_size 
Passed to argument 
bar_thickness 
Passed to argument 
ranking_nterms_max 
Maximum submodel size (number of predictor terms)
for which the predictor names and the corresponding ranking proportions are
added on the xaxis. Using 
ranking_abbreviate 
A single logical value indicating whether the
predictor names in the fulldata predictor ranking should be abbreviated by

ranking_abbreviate_args 
A 
ranking_repel 
Either 
ranking_repel_args 
A 
ranking_colored 
A single logical value indicating whether the points
and the uncertainty bars should be gradientcolored according to the CV
ranking proportions ( 
cumulate 
Passed to argument 
text_angle 
Passed to argument 
... 
Arguments passed to the internal function which is used for
bootstrapping (if applicable; see argument 
The stats
options "mse"
and "rmse"
are only available for:
the traditional projection,
the latent projection with resp_oscale = FALSE
,
the latent projection with resp_oscale = TRUE
in combination with
<refmodel>$family$cats
being NULL
.
The stats
option "acc"
(= "pctcorr"
) is only available for:
the binomial()
family in case of the traditional projection,
all families in case of the augmenteddata projection,
the binomial()
family (on the original response scale) in case of the
latent projection with resp_oscale = TRUE
in combination with
<refmodel>$family$cats
being NULL
,
all families (on the original response scale) in case of the latent
projection with resp_oscale = TRUE
in combination with
<refmodel>$family$cats
being not NULL
.
The stats
option "auc"
is only available for:
the binomial()
family in case of the traditional projection,
the binomial()
family (on the original response scale) in case of the
latent projection with resp_oscale = TRUE
in combination with
<refmodel>$family$cats
being NULL
.
A ggplot2 plotting object (of class gg
and ggplot
). If
ranking_abbreviate
is TRUE
, the output of abbreviate()
is stored in
an attribute called projpred_ranking_abbreviated
(to allow the
abbreviations to be easily mapped back to the original predictor names).
As long as the reference model's performance is computable, it is always
shown in the plot as a dashed red horizontal line. If baseline = "best"
,
the baseline model's performance is shown as a dotted black horizontal line.
If !is.na(thres_elpd)
and any(stats %in% c("elpd", "mlpd"))
, the value
supplied to thres_elpd
(which is automatically adapted internally in case
of the MLPD or deltas = FALSE
) is shown as a dotdashed gray horizontal
line for the reference model and, if baseline = "best"
, as a longdashed
green horizontal line for the baseline model.
# Data:
dat_gauss < data.frame(y = df_gaussian$y, df_gaussian$x)
# The "stanreg" fit which will be used as the reference model (with small
# values for `chains` and `iter`, but only for technical reasons in this
# example; this is not recommended in general):
fit < rstanarm::stan_glm(
y ~ X1 + X2 + X3 + X4 + X5, family = gaussian(), data = dat_gauss,
QR = TRUE, chains = 2, iter = 500, refresh = 0, seed = 9876
)
# Run varsel() (here without crossvalidation, with L1 search, and with small
# values for `nterms_max` and `nclusters_pred`, but only for the sake of
# speed in this example; this is not recommended in general):
vs < varsel(fit, method = "L1", nterms_max = 3, nclusters_pred = 10,
seed = 5555)
print(plot(vs))
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.