porridge-package: Ridge-Type Penalized Estimation of a Potpourri of Models.

In porridge: Ridge-Type Penalized Estimation of a Potpourri of Models

Ridge-Type Penalized Estimation of a Potpourri of Models.

Description

The following functions facilitate the ridge-type penalized estimation of various models. Currently, it includes:

• Generalized ridge estimation of the precision matrix of a Gaussian graphical model (van Wieringen, 2019) through the function ridgePgen. This function is complemented by the functions ridgePgen.kCV, ridgePgen.kCV.banded, ridgePgen.kCV.groups, optPenaltyPgen.kCVauto.banded and optPenaltyPgen.kCVauto.groups for penalty parameters selection through K-fold cross-validation assuming a particularly structured precision matrix.

• Multi-targeted ridge estimation of the precision matrix of a Gaussian graphical model (van Wieringen et al., 2020) through the functions ridgePmultiT. This function is complemented by the functions optPenaltyPmultiT.kCVauto for penalty parameters selection through K-fold cross-validation.

• Gaussian graphical model estimation from data with replicates in ridge penalized fashion (van Wieringen, Chen, 2021) (ridgePrep and ridgePrepEdiag). The two functions optPenaltyPrep.kCVauto and optPenaltyPrepEdiag.kCVauto implement the corresponding K-fold cross-validation procedures for an optimal choice of the penalty parameter.

• Ridge penalized estimation of a mixture of Gaussian graphical models: ridgeGGMmixture and its penalty selection via K-fold cross-validation optPenaltyGGMmixture.kCVauto.

• Targeted and multi-targeted ridge estimation of the regression parameter of the generalized linear model (van Wieringen, Binder, 2022; van Wieringen, 2021; Lettink et al., 2022) through the functions ridgeGLM and ridgeGLMmultiT. This function is complemented by the functions optPenaltyGLM.kCVauto and optPenaltyGLMmultiT.kCVauto for penalty parameters selection through K-fold cross-validation, and the ridgeGLMdof-function for the evaluation of the fitted model's degrees of freedom.

Future versions aim to include more ridge-type functionality.

In part the porridge-package extends/builds upon the rags2ridges- and ragt2ridges-packages, in which some or all functionality of the porridge-package may be absorped at some point in the future.

Author(s)

Wessel N. van Wieringen <[e-mail]>

References

Aflakparast, M., de Gunst, M.C.M., van Wieringen, W.N. (2018), "Reconstruction of molecular network evolution from cross-sectional omics data", Biometrical Journal, 60(3), 547-563.

Lettink, A., Chinapaw, M.J.M., van Wieringen, W.N. (2022), "Two-dimensional fused targeted ridge regression for health indicator prediction from accelerometer data", submitted.

Peeters, C.F.W., Bilgrau, A.E., and van Wieringen, W.N. (2021), "rags2ridges: Ridge Estimation of Precision Matrices from High-Dimensional Data", R package version 2.2.5. https://CRAN.R-project.org/package=rags2ridges.

van Wieringen, W.N. (2020), "ragt2ridges: Ridge Estimation of Vector Auto-Regressive (VAR) Processes", R package version 0.3.4. https://CRAN.R-project.org/package=ragt2ridges.

van Wieringen, W.N. (2019), "The generalized ridge estimator of the inverse covariance matrix", Journal of Computational and Graphical Statistics, 28(4), 932-942.

van Wieringen, W.N. (2021), "Lecture notes on ridge regression", Arxiv preprint, arXiv:1509.09169.

van Wieringen W.N., Chen, Y. (2021), "Penalized estimation of the Gaussian graphical model from data with replicates", Statistics in Medicine, 40(19), 4279-4293.

van Wieringen, W.N., Stam, K.A., Peeters, C.F.W., van de Wiel, M.A. (2020), "Updating of the Gaussian graphical model through targeted penalized estimation", Journal of Multivariate Analysis, 178, Article 104621.

van Wieringen, W.N. Binder, H. (2022), "Sequential learning of regression models by penalized estimation", Journal of Computational and Graphical Statistics, accepted.

See Also

The porridge-package.

porridge documentation built on Oct. 16, 2023, 1:06 a.m.