This function can suggest an appropriate submodel size based on a decision
rule described in section "Details" below. Note that this decision is quite
heuristic and should be interpreted with caution. It is recommended to
examine the results via plot.vsel(), cv_proportions(),
plot.cv_proportions(), and/or summary.vsel() and to make the final
decision based on what is most appropriate for the problem at hand.
suggest_size(object, ...)
# S3 method for vsel
suggest_size(
object,
stat = "elpd",
pct = 0,
type = "upper",
thres_elpd = NA,
warnings = TRUE,
...
)An object of class vsel (returned by varsel() or
cv_varsel()).
Arguments passed to summary.vsel(), except for object, stats
(which is set to stat), type, and deltas (which is set to TRUE).
See section "Details" below for some important arguments which may be
passed here.
Performance statistic (i.e., utility or loss) used for the
decision. See argument stats of summary.vsel() for possible choices.
A number giving the proportion (not percents) of the relative null model utility one is willing to sacrifice. See section "Details" below for more information.
Either "upper" or "lower" determining whether the decision is
based on the upper or lower confidence interval bound, respectively. See
section "Details" below for more information.
Only relevant if stat %in% c("elpd", "mlpd", "gmpd")).
The threshold for the ELPD difference (taking the submodel's ELPD minus the
baseline model's ELPD) above which the submodel's ELPD is considered to be
close enough to the baseline model's ELPD. An equivalent rule is applied in
case of the MLPD and the GMPD. See section "Details" for a formalization.
Supplying NA deactivates this.
Mainly for internal use. A single logical value indicating
whether to throw warnings if automatic suggestion fails. Usually there is
no reason to set this to FALSE.
A single numeric value, giving the suggested submodel size (or NA if the suggestion failed).
The intercept is not counted by suggest_size(), so a suggested size of
zero stands for the intercept-only model.
In general (beware of special cases below), the suggested model
size is the smallest model size \(j \in \{0, 1, ...,
\texttt{nterms\_max}\}\) for which either the
lower or upper bound (depending on argument type) of the
normal-approximation (or bootstrap or exponentiated normal-approximation;
see argument stat) confidence interval (with nominal coverage 1 - alpha; see argument alpha of summary.vsel()) for \(U_j -
U_{\mathrm{base}}\) (with \(U_j\) denoting the \(j\)-th
submodel's true utility and \(U_{\mathrm{base}}\) denoting the
baseline model's true utility)
falls above (or is equal to) $$\texttt{pct} \cdot (u_0 -
u_{\mathrm{base}})$$ where \(u_0\) denotes the null
model's estimated utility and \(u_{\mathrm{base}}\) the baseline
model's estimated utility. The baseline model is either the reference model
or the best submodel found (see argument baseline of summary.vsel()).
In doing so, loss statistics like the root mean squared error (RMSE) and
the mean squared error (MSE) are converted to utilities by multiplying them
by -1, so a call such as suggest_size(object, stat = "rmse", type = "upper") finds the smallest model size whose upper confidence interval
bound for the negative RMSE or MSE exceeds (or is equal to) the cutoff
(or, equivalently, has the lower confidence interval bound for the RMSE or
MSE below---or equal to---the cutoff). This is done to make the
interpretation of argument type the same regardless of argument stat.
For the geometric mean predictive density (GMPD), the decision rule above
is applied on log() scale. In other words, if the true GMPD is denoted by
\(U^\ast_j\) for the \(j\)-th submodel and
\(U^\ast_{\mathrm{base}}\) for the baseline model (so that
\(U_j\) and \(U_{\mathrm{base}}\) from above are given by
\(U_j = \log(U^\ast_j)\) and
\(U_{\mathrm{base}} = \log(U^\ast_{\mathrm{base}})\)), then suggest_size() yields the smallest model size whose
lower or upper (depending on argument type) confidence interval bound for
\(\frac{U^\ast_j}{U^\ast_{\mathrm{base}}}\) exceeds (or
is equal to)
$$(\frac{u^\ast_0}{u^\ast_{\mathrm{base}}})^{\texttt{pct}}$$ where \(u^\ast_0\) denotes the null
model's estimated GMPD and \(u^\ast_{\mathrm{base}}\) the
baseline model's estimated GMPD.
If !is.na(thres_elpd) and stat = "elpd", the decision rule above is
extended: The suggested model size is then the smallest model size \(j\)
fulfilling the rule above or \(u_j - u_{\mathrm{base}} >
\texttt{thres\_elpd}\). Correspondingly, in case
of stat = "mlpd" (and !is.na(thres_elpd)), the suggested model size is
the smallest model size \(j\) fulfilling the rule above or \(u_j -
u_{\mathrm{base}} > \frac{\texttt{thres\_elpd}}{N}\) with \(N\) denoting the number of observations.
Correspondingly, in case of stat = "gmpd" (and !is.na(thres_elpd)), the
suggested model size is the smallest model size \(j\) fulfilling the rule
above or \(\frac{u^\ast_j}{u^\ast_{\mathrm{base}}} >
\exp(\frac{\texttt{thres\_elpd}}{N})\).
For example (disregarding the special extensions in case of
!is.na(thres_elpd) with stat %in% c("elpd", "mlpd", "gmpd")), alpha = 2 * pnorm(-1), pct = 0, and type = "upper" means that we select the
smallest model size for which the upper bound of the 1 - 2 * pnorm(-1)
(approximately 68.3%) confidence interval for \(U_j -
U_{\mathrm{base}}\)
(\(\frac{U^\ast_j}{U^\ast_{\mathrm{base}}}\) in case of
the GMPD) exceeds (or is equal to) zero (one in case of the GMPD), that is
(if stat is a performance statistic for which the normal approximation is
used, not the bootstrap and not the exponentiated normal approximation),
for which the submodel's utility estimate is at most one standard error
smaller than the baseline model's utility estimate (with that standard
error referring to the utility difference).
Apart from the two summary.vsel() arguments mentioned above (alpha and
baseline), resp_oscale is another important summary.vsel() argument
that may be passed via ....
# 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 cross-validation, 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, verbose = FALSE)
print(suggest_size(vs))
#> [1] 3