Complexity selection helpers for penalized backends

Wrapper functions for choosing the penalized tuning parameter used by fit.penalized, fit.LASSO, and fit.fusedLASSO.

Usage

complexity.cv.penalized(response, x, full.data, ...)
complexity.LASSO(response, x, full.data, ...)
complexity.fusedLASSO(response, x, full.data, ...)

Arguments

response

response used for model fitting. Survival responses may be passed as a Surv object or as a two-column matrix with time and status.

x

n * p matrix of covariates.

full.data

data frame containing the full data set. This argument is accepted to match the peperr complexity-function contract.

...

additional arguments passed to profL1.

Value

A scalar tuning value, typically the selected lambda1.

Details

complexity.cv.penalized profiles the cross-validated lambda1 values returned by profL1 and keeps the maximizer. complexity.LASSO is a compatibility alias. complexity.fusedLASSO performs the same search while forwarding fusedl = TRUE.

See also

fit.penalized, fit.LASSO, fit.fusedLASSO, profL1