| Type: | Package |
| Title: | Broken Adaptive Ridge Regression with Cyclops |
| Version: | 1.0.0 |
| Date: | 2023-04-05 |
| Maintainer: | Marc A. Suchard <msuchard@ucla.edu> |
| Description: | Approximates best-subset selection (L0) regression with an iteratively adaptive Ridge (L2) penalty for large-scale models. This package uses Cyclops for an efficient implementation and the iterative method is described in Kawaguchi et al (2020) <doi:10.1002/sim.8438> and Li et al (2021) <doi:10.1016/j.jspi.2020.12.001>. |
| License: | Apache License 2.0 |
| Depends: | R (≥ 3.2.2), Cyclops (≥ 3.0.0) |
| Imports: | futile.logger, bit64 |
| Suggests: | testthat, survival, knitr, rmarkdown |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.1.2 |
| NeedsCompilation: | no |
| Packaged: | 2023-04-06 12:24:15 UTC; msuchard |
| Author: | Marc A. Suchard [aut, cre], Eric Kawaguchi [aut], Ning Li [aut], Gang Li [aut], Observational Health Data Sciences and Informatics [cph] |
| Repository: | CRAN |
| Date/Publication: | 2023-04-07 08:50:10 UTC |
Create a BAR Cyclops prior object
Description
createBarPrior creates a BAR Cyclops prior object for use with fitCyclopsModel.
Usage
createBarPrior(
penalty = "bic",
exclude = c(),
forceIntercept = FALSE,
fitBestSubset = FALSE,
initialRidgeVariance = 10000,
tolerance = 1e-08,
maxIterations = 10000,
threshold = 1e-06,
delta = 0
)
Arguments
penalty |
Specifies the BAR penalty; possible values are 'BIC' or 'AIC' or a numeric value |
exclude |
A vector of numbers or covariateId names to exclude from prior |
forceIntercept |
Logical: Force intercept coefficient into regularization |
fitBestSubset |
Logical: Fit final subset with no regularization |
initialRidgeVariance |
Numeric: variance used for algorithm initiation |
tolerance |
Numeric: maximum abs change in coefficient estimates from successive iterations to achieve convergence |
maxIterations |
Numeric: maxium iterations to achieve convergence |
threshold |
Numeric: absolute threshold at which to force coefficient to 0 |
delta |
Numeric: change from 2 in ridge norm dimension |
Value
A BAR Cyclops prior object of class inheriting from
"cyclopsPrior" for use with fitCyclopsModel.
Examples
prior <- createBarPrior(penalty = "bic")
Create a fastBAR Cyclops prior object
Description
createFastBarPrior creates a fastBAR Cyclops prior object for use with fitCyclopsModel.
Usage
createFastBarPrior(
penalty = 0,
exclude = c(),
forceIntercept = FALSE,
fitBestSubset = FALSE,
initialRidgeVariance = 10000,
tolerance = 1e-08,
maxIterations = 10000,
threshold = 1e-06
)
Arguments
penalty |
Specifies the BAR penalty |
exclude |
A vector of numbers or covariateId names to exclude from prior |
forceIntercept |
Logical: Force intercept coefficient into regularization |
fitBestSubset |
Logical: Fit final subset with no regularization |
initialRidgeVariance |
Numeric: variance used for algorithm initiation |
tolerance |
Numeric: maximum abs change in coefficient estimates from successive iterations to achieve convergence |
maxIterations |
Numeric: maximum iterations to achieve convergence |
threshold |
Numeric: absolute threshold at which to force coefficient to 0 |
Value
A BAR Cyclops prior object of class inheriting from
"cyclopsPrior" for use with fitCyclopsModel.
Examples
nobs = 500; ncovs = 100
prior <- createFastBarPrior(penalty = log(ncovs), initialRidgeVariance = 1 / log(ncovs))