ecpc-package {ecpc}

Flexible Co-Data Learning for High-Dimensional Prediction

Description

Fit linear, logistic and Cox survival regression models penalised with adaptive multi-group ridge penalties. The multi-group penalties correspond to groups of covariates defined by (multiple) co-data sources. Group hyperparameters are estimated with an empirical Bayes method of moments, penalised with an extra level of hyper shrinkage. Various types of hyper shrinkage may be used for various co-data. Co-data may be continuous or categorical. The method accommodates inclusion of unpenalised covariates, posterior selection of covariates and multiple data types. The model fit is used to predict for new samples. The name 'ecpc' stands for Empirical Bayes, Co-data learnt, Prediction and Covariate selection. See Van Nee et al. (2020) <arXiv:2005.04010>.

Details

The DESCRIPTION file:

Package:	ecpc
Type:	Package
Title:	Flexible Co-Data Learning for High-Dimensional Prediction
Version:	3.1.1
Date:	2023-02-27
Authors@R:	c(person(c("Mirrelijn","M."), "van Nee", role = c("aut", "cre"), email = "[e-mail]"), person(c("Lodewyk","F.A."), "Wessels", role = "aut"), person(c("Mark","A."), "van de Wiel", role = "aut"))
Author:	Mirrelijn M. van Nee [aut, cre], Lodewyk F.A. Wessels [aut], Mark A. van de Wiel [aut]
Maintainer:	Mirrelijn M. van Nee <[e-mail]>
Depends:	R (>= 3.5.0)
Imports:	glmnet, stats, Matrix, gglasso, mvtnorm, CVXR, multiridge (>= 1.5), survival, pROC, mgcv, pracma, JOPS, quadprog, checkmate
Suggests:	Rsolnp, expm, foreach, doParallel, parallel, ggplot2, ggraph, igraph, ggpubr, scales, dplyr, magrittr, nnls
Description:	Fit linear, logistic and Cox survival regression models penalised with adaptive multi-group ridge penalties. The multi-group penalties correspond to groups of covariates defined by (multiple) co-data sources. Group hyperparameters are estimated with an empirical Bayes method of moments, penalised with an extra level of hyper shrinkage. Various types of hyper shrinkage may be used for various co-data. Co-data may be continuous or categorical. The method accommodates inclusion of unpenalised covariates, posterior selection of covariates and multiple data types. The model fit is used to predict for new samples. The name 'ecpc' stands for Empirical Bayes, Co-data learnt, Prediction and Covariate selection. See Van Nee et al. (2020) <arXiv:2005.04010>.
License:	GPL (>= 3)
URL:	http://dx.doi.org/10.1002/sim.9162
RoxygenNote:	7.2.0

Index of help topics:

coef.ecpc               Obtain coefficients from 'ecpc' object
createCon               Create a list of constraints for co-data weight
                        estimation
createGroupset          Create a group set (groups) of variables
createS                 Create a generalised penalty matrix
createZforGroupset      Create a co-data matrix Z for a group set
createZforSplines       Create a co-data matrix Z of splines
cv.ecpc                 Cross-validation for 'ecpc'
ecpc                    Fit adaptive multi-group ridge GLM with
                        hypershrinkage
ecpc-package            Flexible Co-Data Learning for High-Dimensional
                        Prediction
hierarchicalLasso       Fit hierarchical lasso using LOG penalty
obtainHierarchy         Obtain hierarchy
plot.ecpc               Plot an 'ecpc' object
postSelect              Perform posterior selection
predict.ecpc            Predict for new samples for 'ecpc' object
print.ecpc              Print summary of 'ecpc' object
produceFolds            Produce folds
simDat                  Simulate data
splitMedian             Discretise continuous data in multiple
                        granularities
visualiseGroupset       Visualise a group set
visualiseGroupsetweights
                        Visualise estimated group set weights
visualiseGroupweights   Visualise estimated group weights

See ecpc for example code.

Author(s)

Mirrelijn M. van Nee [aut, cre], Lodewyk F.A. Wessels [aut], Mark A. van de Wiel [aut]

Maintainer: Mirrelijn M. van Nee <[e-mail]>