R packages by hlorenzo

ddsPLS - Data-Driven Sparse Partial Least Squares

A sparse Partial Least Squares implementation which uses soft-threshold estimation of the covariance matrices and therein introduces sparsity. Number of components and regularization coefficients are automatically set.

Last updated 1 years ago

missing-datamulti-blockplssupervised-learningsvdvariable-selectioncpp

3.70 score 7 scripts 297 downloads

SEPaLS - Shrinkage for Extreme Partial Least-Squares (SEPaLS)

Regression context for the Partial Least Squares framework for Extreme values. Estimations of the Shrinkage for Extreme Partial Least-Squares (SEPaLS) estimators, an adaptation of the original Partial Least Squares (PLS) method tailored to the extreme-value framework. The SEPaLS project is a joint work by Stephane Girard, Hadrien Lorenzo and Julyan Arbel. R code to replicate the results of the paper is available at <https://github.com/hlorenzo/SEPaLS_simus>. Extremes within PLS was already studied by one of the authors, see M Bousebeta, G Enjolras, S Girard (2023) <doi:10.1016/j.jmva.2022.105101>.

Last updated 1 years ago

2.70 score 203 downloads