frs: Forward Regression Selection Framework

In pboost: Profile Boosting Framework for Parametric Models

Forward Regression Selection Framework

Description

frs() is a generic workhorse function of forward regression selection for parametric regression.

Usage

frs(
  xmat,
  yvec,
  fitFun,
  ...,
  use.formula = TRUE,
  use.intercept = TRUE,
  selectFun = logLik,
  stopFun = "EBIC",
  keep = NULL,
  maxK = NULL,
  verbose = FALSE
)

Arguments

xmat	See pboost.
yvec	See pboost.
fitFun	See pboost.
...	See pboost.
use.formula	See pboost.
use.intercept	See pboost.
selectFun	A function to evaluate the importance of an unselected feature when it is added to current model. The default is logLik, meaning that the feature with the largest post-added log-likelihood is identified as the next one to be added to the model. Note that selectFun is only used for selecting features, and it does not affect the stopping rule of forward regression selection, which is determined by stopFun.
stopFun	See pboost.
keep	See pboost.
maxK	See pboost.
verbose	See pboost.

Value

Model object fitted on the selected features.

See Also

fbetareg, fcoxph, fglm, flm, frq, fsar.

pboost documentation built on May 24, 2026, 9:08 a.m.